fix: decouple session resume from isFirstVisit guard

When frontmatter validation fails, the step is never written to CAS, so
isFirstVisit remains true on the next run.  Both agent-claude-code and
agent-hermes gated session cache lookup behind !isFirstVisit, which
caused them to start a fresh session (and a new worktree) instead of
resuming the one that already has all the work done.

Remove the isFirstVisit guard from both adapters so they always check
the session cache.  If no cache entry exists, the behavior is unchanged
(fresh session).  If a cache entry exists, the agent resumes the
previous session regardless of CAS step state.

Fixes #139

.changeset/start-new-resume.md

@@ -1,9 +0,0 @@
----
-"@united-workforce/cli": minor
-"@united-workforce/util": patch
----
-
-feat: replace $START `_` status with `new`/`resume` semantics
-
-BREAKING: All workflow YAML files must update `$START._` to `$START.new` + `$START.resume`.
-The `resume` edge prompt replaces the previously hardcoded resume message in the CLI.

.workflows/solve-issue.yaml

@@ -1,247 +0,0 @@
-name: "solve-issue"
-description: "TDD-driven issue resolution for small, focused changes. Loop protection relies on engine maxRounds."
-roles:
-  planner:
-    description: "Analyzes issue and outputs a TDD test spec"
-    goal: "You are a planning agent. You analyze Gitea issues and produce a TDD test specification that downstream roles will implement and verify."
-    capabilities:
-      - issue-analysis
-      - planning
-    procedure: |
-      On first run (no previous steps):
-      1. Read the issue and all comments from Gitea using `tea issues <number> -r <owner/repo>`
-      2. Look for project conventions files (CLAUDE.md, CONTRIBUTING.md, .cursor/rules/) in the repo
-      3. Assess whether the issue has enough information to produce a test spec
-      4. If insufficient info: comment on the issue via `echo "..." | tea comment <number> -r <owner/repo>` (skip if you already commented), then output $status=insufficient_info
-      5. If sufficient: produce a detailed TDD test spec in markdown covering all scenarios
-
-      On subsequent runs (bounced back by tester with fix_spec):
-      1. Read the tester's output from the previous step to understand what's wrong with the spec
-      2. Revise the test spec accordingly
-
-      After producing the test spec:
-      1. The test spec is stored in CAS automatically by the uwf pipeline (agents do not need to call `ocas put` directly)
-      2. Put the plan hash in frontmatter.plan (required when $status=ready)
-      3. Set repoPath to the absolute path of the repository root
-
-      IMPORTANT: Extract the repo remote (owner/repo) from git:
-      ```bash
-      git remote get-url origin | sed 's|.*[:/]\([^/]*/[^.]*\).*|\1|'
-      ```
-      Store the result as repoRemote in your frontmatter output so downstream roles can use it for tea/API calls.
-    output: "Output a brief summary of the test spec. Set $status to ready (with plan hash and repoPath) or insufficient_info."
-    frontmatter:
-      oneOf:
-        - properties:
-            $status: { const: "ready" }
-            plan: { type: string }
-            repoPath: { type: string }
-            repoRemote: { type: string }
-          required: [$status, plan, repoPath, repoRemote]
-        - properties:
-            $status: { const: "insufficient_info" }
-            reason: { type: string }
-          required: [$status, reason]
-  developer:
-    description: "TDD implementation per test spec"
-    goal: "You are a developer agent. You implement code changes following TDD — write tests first, then implementation."
-    capabilities:
-      - coding
-    procedure: |
-      IMPORTANT: Always work in a git worktree, NEVER modify the main working directory directly.
-      The repo path and other details are provided in your task prompt.
-
-      Before starting any work, set up an isolated worktree:
-      1. cd into the repo path provided in your task prompt
-      2. `git fetch origin` to get latest refs
-      3. First time (no existing branch):
-         - `git worktree add .worktrees/fix/<issue-number>-<short-slug> -b fix/<issue-number>-<short-slug> origin/main`
-         - `cd .worktrees/fix/<issue-number>-<short-slug> && bun install`
-      4. If bounced back from reviewer or tester (branch already exists):
-         - cd into the existing worktree under `.worktrees/fix/<issue-number>-<short-slug>`
-         - `git fetch origin && git rebase origin/main`
-      5. ALL subsequent work must happen inside the worktree directory.
-
-      Then implement TDD:
-      6. Read the test spec from CAS: `ocas get <plan hash>` (find the hash from the planner's output in your task prompt)
-      7. If bounced back from reviewer or tester: read the previous role's feedback in your task prompt
-      8. Write tests first based on the spec
-      9. Implement the code to make tests pass
-      10. Ensure `bun run build` passes with no errors
-      11. Run `bun test` to verify all tests pass
-          - If tests fail on first run:
-            * Read the test output carefully for missing imports or setup issues
-            * Check if you're running tests from the correct working directory (package root vs workspace root)
-            * Fix the immediate issue and rerun ONCE
-            * If tests still fail after 2 attempts: check the test spec for ambiguities
-            * If stuck after 3 test cycles: set $status=failed with detailed error report rather than continuing blind retries
-      12. MANDATORY VERIFICATION before reporting done:
-          - Run `git branch --show-current` and confirm branch name matches expected
-          - Run `git status` and verify changed files exist
-          - Run `ls -la <key-implementation-files>` to verify they exist on disk
-          - If ANY verification fails: retry the implementation, do NOT report done
-
-      If you cannot complete the implementation (e.g. the issue is too complex, blocked by external factors,
-      or repeated attempts fail), set $status=failed with a reason.
-    output: "List all files changed and provide a summary. Set $status to done (with branch/worktree), or failed (with reason)."
-    frontmatter:
-      oneOf:
-        - properties:
-            $status: { const: "done" }
-            branch: { type: string }
-            worktree: { type: string }
-            repoRemote: { type: string }
-          required: [$status, branch, worktree]
-        - properties:
-            $status: { const: "failed" }
-            reason: { type: string }
-          required: [$status, reason]
-  reviewer:
-    description: "Code standards compliance check"
-    goal: "You are a code reviewer. You verify code standards compliance — NOT functionality (that's the tester's job)."
-    capabilities:
-      - code-review
-      - static-analysis
-    procedure: |
-      The worktree path is provided in your task prompt. cd into it first.
-
-      CRITICAL: You MUST execute every verification command below. Do NOT report results without running the actual commands. Do NOT rely on prior context or assumptions.
-
-      Before reviewing, verify the worktree and branch exist:
-      0. Run `cd <worktree-path> && pwd` to confirm the path is accessible
-         - If the cd fails: the worktree truly doesn't exist, reject with that reason
-         - If the cd succeeds: proceed with step 1 below
-      1. Run `git branch --show-current` — confirm the branch name references the issue number being worked on
-      2. If the branch doesn't correspond to the issue, flag it in your output and reject
-
-      Then perform code review:
-      Hard checks (must all pass):
-      3. `bun run build` — no build errors
-      4. `bunx biome check` — no lint violations
-      5. TypeScript strict mode — no type errors
-
-      Soft checks (review against project conventions if CLAUDE.md / .cursor/rules exist):
-      - Naming conventions, module boundaries, code style
-      - No `console.log` in production code
-      - No dynamic imports in production code
-
-      Only review standards compliance. Do NOT test functionality.
-      If rejecting, you MUST explain the specific reason in your output.
-    output: "Explain your decision with specific file/line references. Set $status to approved (with branch/worktree) or rejected (with comments)."
-    frontmatter:
-      oneOf:
-        - properties:
-            $status: { const: "approved" }
-            branch: { type: string }
-            worktree: { type: string }
-            repoRemote: { type: string }
-          required: [$status, branch, worktree]
-        - properties:
-            $status: { const: "rejected" }
-            comments: { type: string }
-            worktree: { type: string }
-            repoRemote: { type: string }
-          required: [$status, comments, worktree]
-  tester:
-    description: "Functional correctness verification"
-    goal: "You are a tester agent. You verify that the implementation correctly satisfies every scenario in the test spec."
-    capabilities:
-      - testing
-    procedure: |
-      The worktree path is provided in your task prompt. cd into it first.
-
-      1. Run `bun test` for automated test verification
-      2. Read the test spec from CAS: `ocas get <plan hash>` (find the hash from the planner step in the thread history)
-      3. Verify each scenario in the spec is covered and passing
-      4. Determine outcome:
-         - passed: all scenarios verified, tests pass
-         - fix_code: tests fail or implementation doesn't match spec → send back to developer
-         - fix_spec: the spec itself is wrong or incomplete → send back to planner
-    output: "Report test results per scenario. Set $status to passed (with branch/worktree), fix_code (with report), or fix_spec (with report)."
-    frontmatter:
-      oneOf:
-        - properties:
-            $status: { const: "passed" }
-            branch: { type: string }
-            worktree: { type: string }
-            repoRemote: { type: string }
-          required: [$status, branch, worktree]
-        - properties:
-            $status: { const: "fix_code" }
-            report: { type: string }
-            repoRemote: { type: string }
-            worktree: { type: string }
-            branch: { type: string }
-          required: [$status, report]
-        - properties:
-            $status: { const: "fix_spec" }
-            report: { type: string }
-            repoRemote: { type: string }
-            worktree: { type: string }
-            branch: { type: string }
-          required: [$status, report]
-  committer:
-    description: "Commits and creates PR"
-    goal: "You are a committer agent. You create a clean commit and push a PR linking the original issue."
-    capabilities: []
-    procedure: |
-      The worktree path, branch name, and repo remote (owner/repo) are provided in your task prompt.
-      cd into the worktree first.
-
-      Note: You inherit the developer's worktree and branch. Do NOT create a new branch.
-      1. Check `git status` — if working tree is clean and branch is ahead of origin, skip to step 3 (push).
-      2. If there are unstaged/uncommitted changes: `git add -A` then `git commit -m "type: description\n\nFixes #N"`
-      3. Push the branch: `git push -u origin <branch-name>`
-      4. **Verify push succeeded** — run `git ls-remote origin <branch-name>` and confirm it prints a commit hash.
-         - If no output or push failed: capture the error, mark hook_failed
-      5. Create a PR using the Gitea API (do NOT use `tea pr create` — it fails in worktrees):
-         ```bash
-         GITEA_TOKEN=$(cfg get GITEA_TOKEN)
-         curl -s -X POST -H "Authorization: token $GITEA_TOKEN" -H "Content-Type: application/json" \
-           "https://git.shazhou.work/api/v1/repos/<owner>/<repo>/pulls" \
-           -d '{"title":"...","body":"...","head":"<branch>","base":"main"}'
-         ```
-         - The repo remote (owner/repo format, e.g. "shazhou/united-workforce") is given in your task prompt — use it directly.
-         - PR body must include: What / Why / Changes / Ref sections, with `Fixes #N` in Ref
-      6. **Verify PR was created** — parse the curl response JSON: it must contain a `"number"` field. Print the PR URL.
-         - If curl returns an error or no number field: capture the response, mark hook_failed
-      7. After PR creation, clean up the worktree:
-         - cd to the repo root (parent of .worktrees)
-         - `git worktree remove <worktree-path>`
-    output: "Include PR URL on success or error log on failure. Set $status to committed (with prUrl) or hook_failed (with error)."
-    frontmatter:
-      oneOf:
-        - properties:
-            $status: { const: "committed" }
-            prUrl: { type: string }
-            repoRemote: { type: string }
-            worktree: { type: string }
-            branch: { type: string }
-          required: [$status, prUrl]
-        - properties:
-            $status: { const: "hook_failed" }
-            error: { type: string }
-            repoRemote: { type: string }
-            worktree: { type: string }
-            branch: { type: string }
-          required: [$status, error]
-graph:
-  $START:
-    new: { role: "planner", prompt: "Analyze the issue and produce an implementation plan." }
-    resume: { role: "planner", prompt: "Review the previous run output and continue the work." }
-  planner:
-    insufficient_info: { role: "$SUSPEND", prompt: "信息不足，需要补充：{{{reason}}}" }
-    ready: { role: "developer", prompt: "Implement the TDD test spec (CAS hash: {{{plan}}}) in repo {{{repoPath}}}. Repo remote: {{{repoRemote}}}." }
-  developer:
-    done: { role: "reviewer", prompt: "Review branch {{{branch}}} at {{{worktree}}} for code standards compliance. Repo remote: {{{repoRemote}}}." }
-    failed: { role: "$END", prompt: "Developer failed: {{{reason}}}. Ending workflow." }
-  reviewer:
-    rejected: { role: "developer", prompt: "Reviewer rejected: {{{comments}}}. Fix the issues in repo {{{worktree}}}. Repo remote: {{{repoRemote}}}." }
-    approved: { role: "tester", prompt: "Review passed. Run tests on branch {{{branch}}} at {{{worktree}}}. Repo remote: {{{repoRemote}}}." }
-  tester:
-    fix_code: { role: "developer", prompt: "Tests found code issues: {{{report}}}. Fix and re-submit. Worktree: {{{worktree}}}. Repo remote: {{{repoRemote}}}." }
-    fix_spec: { role: "planner", prompt: "Tests found spec issues: {{{report}}}. Revise the test spec. Repo remote: {{{repoRemote}}}." }
-    passed: { role: "committer", prompt: "All tests passed. Commit and push branch {{{branch}}} from {{{worktree}}}. Repo remote (owner/repo): {{{repoRemote}}}." }
-  committer:
-    hook_failed: { role: "developer", prompt: "Push hook failed: {{{error}}}. Fix and re-submit. Worktree: {{{worktree}}}. Repo remote: {{{repoRemote}}}." }
-    committed: { role: "$END", prompt: "PR created: {{{prUrl}}}. Workflow complete." }

examples/analyze-topic.yaml

@@ -23,7 +23,7 @@ roles:
       type: object
       properties:
         $status:
-          enum: ["done"]
+          const: done
         thesis:
           type: string
         keyPoints:

examples/debate.yaml

@@ -1,63 +1,131 @@
-name: "debate"
+name: debate
-description: "Structured debate between two sides. Tests cross-process session resume."
+description: "Multi-role structured debate with critical thinking framework and host summary."
+
+# Shared frontmatter schema for debater roles (YAML anchor)
+x-debater-frontmatter: &debater-frontmatter
+  type: object
+  oneOf:
+    - properties:
+        $status: { const: speak }
+        argument: { type: string }
+      required: [$status, argument]
+    - properties:
+        $status: { const: conceded }
+        reason: { type: string }
+      required: [$status, reason]
+    - properties:
+        $status: { const: final }
+        closing: { type: string }
+      required: [$status, closing]
+
 roles:
-  against:
+  proponent:
-    description: "Argues against the proposition"
+    description: "Argues FOR the proposition"
-    goal: |
+    goal: "Build a compelling case for the proposition through logical reasoning and evidence"
-      You are a skilled debater arguing AGAINST the proposition.
+    capabilities: []
-      Be logical, cite evidence, and directly address your opponent's points.
-      Keep each argument concise (under 200 words).
-    capabilities:
-      - argumentation
-      - critical-thinking
     procedure: |
-      1. If this is the opening, present your strongest argument against the proposition.
+      You are an experienced scholar arguing FOR the proposition.
-      2. If responding to the other side, directly counter their points with evidence and logic.
+
-      3. If you find yourself genuinely convinced by the other side, you may concede.
+      ## Critical Thinking Framework (execute before every speech)
-    output: |
+
-      Provide your argument in the frontmatter.
+      ### A. Pre-speech reflection (internal, do not output)
-      Set status to "conceded" ONLY if you are genuinely convinced and wish to stop debating.
+      - Does every step in my argument chain hold? Any hidden assumptions or logical gaps?
-      Otherwise set status to "continue".
+      - If I were my opponent, how would I attack this? Where am I weakest?
+      - Does my evidence actually support my claim, or could it backfire?
+      - Should I go on offense or defense this round?
+
+      ### B. Evidence discipline
+      - Verify key numbers — watch for order-of-magnitude errors
+      - Assess data freshness — fast-moving fields have short half-lives
+      - Distinguish primary data from secondary citations, expert opinion, and common assumptions
+
+      ### C. Anti-fragility
+      - Anticipate counterarguments; preemptively strengthen or strategically abandon weak points
+      - Catch logical gaps, data misuse, or outdated claims in your opponent's reasoning
+
+      ## Rules
+      1. Check Thread Progress to see how many times you have spoken.
+      2. On your 3rd speech, you MUST output $status: final (closing statement).
+      3. If genuinely convinced by the opponent, output $status: conceded.
+      4. Otherwise output $status: speak and counter the opponent's points.
+      5. Be rigorous, cite evidence, stay concise.
+    output: "Debate argument"
+    frontmatter: *debater-frontmatter
+
+  opponent:
+    description: "Argues AGAINST the proposition"
+    goal: "Build a compelling case against the proposition through logical reasoning and evidence"
+    capabilities: []
+    procedure: |
+      You are an experienced scholar arguing AGAINST the proposition.
+
+      ## Critical Thinking Framework (execute before every speech)
+
+      ### A. Pre-speech reflection (internal, do not output)
+      - Does every step in my argument chain hold? Any hidden assumptions or logical gaps?
+      - If I were my opponent, how would I attack this? Where am I weakest?
+      - Does my evidence actually support my claim, or could it backfire?
+      - Should I go on offense or defense this round?
+
+      ### B. Evidence discipline
+      - Verify key numbers — watch for order-of-magnitude errors
+      - Assess data freshness — fast-moving fields have short half-lives
+      - Distinguish primary data from secondary citations, expert opinion, and common assumptions
+
+      ### C. Anti-fragility
+      - Anticipate counterarguments; preemptively strengthen or strategically abandon weak points
+      - Catch logical gaps, data misuse, or outdated claims in your opponent's reasoning
+
+      ## Rules
+      1. Check Thread Progress to see how many times you have spoken.
+      2. On your 3rd speech, or when the proponent has issued a final statement, you MUST output $status: final.
+      3. If genuinely convinced by the proponent, output $status: conceded.
+      4. Otherwise output $status: speak and counter the proponent's points.
+      5. Be rigorous, cite evidence, stay concise.
+    output: "Debate argument"
+    frontmatter: *debater-frontmatter
+
+  host:
+    description: "Debate moderator — delivers impartial summary and verdict"
+    goal: "Objectively review the debate, analyze both sides, and deliver a verdict"
+    capabilities: []
+    procedure: |
+      You are an experienced academic debate moderator.
+
+      ## Task
+      1. Outline each side's core arguments
+      2. Evaluate reasoning quality and evidence use
+      3. Highlight the most impactful exchanges
+      4. Analyze the deeper significance of the topic
+      5. Deliver an overall verdict
+
+      ## Style
+      - Impartial but with independent judgment
+      - Substantive, not superficial
+    output: "Debate summary report"
     frontmatter:
       type: object
       properties:
-        $status:
+        $status: { const: done }
-          enum: ["continue", "conceded"]
+        summary: { type: string }
-        argument:
+        highlights: { type: string }
-          type: string
+        verdict: { type: string }
-      required: [$status, argument]
+      required: [$status, summary, highlights, verdict]
-  for:
+
-    description: "Argues for the proposition"
-    goal: |
-      You are a skilled debater arguing FOR the proposition.
-      Be logical, cite evidence, and directly address your opponent's points.
-      Keep each argument concise (under 200 words).
-    capabilities:
-      - argumentation
-      - critical-thinking
-    procedure: |
-      1. Read the opposing side's latest argument carefully.
-      2. Counter their points with evidence and logic.
-      3. If you find yourself genuinely convinced by the other side, you may concede.
-    output: |
-      Provide your argument in the frontmatter.
-      Set status to "conceded" ONLY if you are genuinely convinced and wish to stop debating.
-      Otherwise set status to "continue".
-    frontmatter:
-      type: object
-      properties:
-        $status:
-          enum: ["continue", "conceded"]
-        argument:
-          type: string
-      required: [$status, argument]
 graph:
   $START:
-    new: { role: "against", prompt: "Present your opening argument against the proposition." }
+    new: { role: proponent, prompt: "The debate begins. You are arguing FOR the proposition. Present your opening argument." }
-    resume: { role: "against", prompt: "Review the previous debate output and continue the argument against the proposition." }
+    resume: { role: proponent, prompt: "The debate continues." }
-  against:
+
-    conceded: { role: "$END", prompt: "The against side conceded. Debate over." }
+  proponent:
-    continue: { role: "for", prompt: "Counter the opposing argument: {{{argument}}}" }
+    speak: { role: opponent, prompt: "Proponent argues:\n\n{{{argument}}}\n\nYou are the opponent. Counter this argument." }
-  for:
+    conceded: { role: host, prompt: "The proponent conceded: {{{reason}}}\n\nPlease summarize the debate." }
-    conceded: { role: "$END", prompt: "The for side conceded. Debate over." }
+    final: { role: opponent, prompt: "Proponent's closing statement:\n\n{{{closing}}}\n\nYou are the opponent. Deliver your final response." }
-    continue: { role: "against", prompt: "Counter the opposing argument: {{{argument}}}" }
+
+  opponent:
+    speak: { role: proponent, prompt: "Opponent argues:\n\n{{{argument}}}\n\nYou are the proponent. Counter this argument." }
+    conceded: { role: host, prompt: "The opponent conceded: {{{reason}}}\n\nPlease summarize the debate." }
+    final: { role: host, prompt: "Opponent's closing statement:\n\n{{{closing}}}\n\nThe debate is over. Please summarize." }
+
+  host:
+    done: { role: "$END", prompt: "Summary complete." }

examples/eval-simple.yaml

@@ -18,8 +18,7 @@ roles:
       type: object
       properties:
         $status:
-          type: string
+          const: done
-          enum: [done]
         summary:
           type: string
       required: [$status, summary]

examples/solve-issue.yaml

@@ -1,5 +1,5 @@
 name: "solve-issue"
-description: "TDD-driven issue resolution for small, focused changes. Loop protection relies on engine maxRounds."
+description: "TDD-driven issue resolution for small, focused changes. Loop protection relies on engine maxRounds. Uses pnpm."
 roles:
   planner:
     description: "Analyzes issue and outputs a TDD test spec"
@@ -80,7 +80,7 @@ roles:
       2. `git fetch origin` to get latest refs
       3. First time (no existing branch):
          - `git worktree add .worktrees/fix/<issue-number>-<short-slug> -b fix/<issue-number>-<short-slug> origin/main`
-         - `cd .worktrees/fix/<issue-number>-<short-slug> && bun install`
+         - `cd .worktrees/fix/<issue-number>-<short-slug> && pnpm install`
       4. If continuing on existing branch (prompt says "Continue work on existing branch" or provides a worktree path):
          - cd directly into the worktree path provided in the prompt
          - `git fetch origin && git rebase origin/main`
@@ -95,8 +95,20 @@ roles:
       7. If bounced back from reviewer or tester: read the previous role's feedback in your task prompt
       8. Write tests first based on the spec
       9. Implement the code to make tests pass
-      10. Ensure `bun run build` passes with no errors
+      10. Ensure `pnpm run build` passes with no errors
-      11. Run `bun test` to verify all tests pass
+      11. Run `pnpm test` to verify all tests pass
+
+      After implementation, before reporting done:
+      12. Add a changeset file (`.changeset/<short-slug>.md`) with correct bump type:
+          - `patch` for bug fixes, internal refactors, test-only changes
+          - `minor` for new features, new CLI commands, new API surfaces
+          - `major` for breaking changes
+          List every affected package in the changeset frontmatter.
+      13. Update documentation if the change affects user-facing behavior:
+          - `README.md` — usage examples, feature descriptions
+          - `.cards/` — architecture decision records (if applicable)
+          - CLI prompt subcommand output (if CLI help text changes)
+          - CLI `--help` text (if flags/commands are added or changed)
 
       If you cannot complete the implementation (e.g. the issue is too complex, blocked by external factors,
       or repeated attempts fail), set $status=failed with a reason.
@@ -127,8 +139,8 @@ roles:
 
       Then perform code review:
       Hard checks (must all pass):
-      3. `bun run build` — no build errors
+      3. `pnpm run build` — no build errors
-      4. `bunx biome check` — no lint violations
+      4. `pnpm run check` — no lint violations
       5. TypeScript strict mode — no type errors
 
       Soft checks (review against project conventions if CLAUDE.md / .cursor/rules exist):
@@ -136,6 +148,14 @@ roles:
       - No `console.log` in production code
       - No dynamic imports in production code
 
+      Documentation & changeset checks:
+      6. Changeset exists in `.changeset/` with correct bump type (`patch`/`minor`/`major`) and lists all affected packages
+      7. If the change is user-facing, documentation is updated:
+         - `README.md` reflects new/changed behavior
+         - `.cards/` architecture cards updated if design decisions changed
+         - CLI prompt subcommand output updated (if it generates skill/reference content)
+         - CLI `--help` text matches new flags/commands
+
       Only review standards compliance. Do NOT test functionality.
       If rejecting, you MUST explain the specific reason in your output.
     output: "Explain your decision with specific file/line references. Set $status to approved (with branch/worktree) or rejected (with comments)."
@@ -159,7 +179,7 @@ roles:
     procedure: |
       The worktree path is provided in your task prompt. cd into it first.
 
-      1. Run `bun test` for automated test verification
+      1. Run `pnpm test` for automated test verification
       2. Read the test spec from CAS: `ocas get <plan hash>` (find the hash from the planner step in the thread history)
       3. Verify each scenario in the spec is covered and passing
       4. Determine outcome:

packages/agent-builtin/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@united-workforce/agent-builtin",
-  "version": "0.1.1",
+  "version": "0.1.2",
   "files": [
     "src",
     "dist",

packages/agent-builtin/src/cli.ts

@@ -1,4 +1,4 @@
-#!/usr/bin/env node
+#!/usr/bin/env -S node --disable-warning=ExperimentalWarning
 
 // eslint-disable-next-line -- dynamic import for version
 const pkg = await import("../package.json", { with: { type: "json" } });

packages/agent-claude-code/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@united-workforce/agent-claude-code",
-  "version": "0.1.1",
+  "version": "0.1.3",
   "files": [
     "src",
     "dist",

packages/agent-claude-code/src/claude-code.ts

@@ -7,6 +7,7 @@ import {
   type AgentRunResult,
   buildContinuationPrompt,
   buildRolePrompt,
+  buildThreadProgress,
   createAgent,
   getCachedSessionId,
   setCachedSessionId,
@@ -27,6 +28,10 @@ export function buildClaudeCodePrompt(ctx: AgentContext): string {
   if (ctx.outputFormatInstruction !== undefined && ctx.outputFormatInstruction !== "") {
     parts.push(ctx.outputFormatInstruction, "");
   }
+
+  // Inject thread progress so the agent knows step count and role visit count
+  parts.push(buildThreadProgress(ctx.steps, ctx.role), "");
+
   parts.push(rolePrompt, "", "## Task", ctx.start.prompt);
 
   if (!ctx.isFirstVisit) {
@@ -171,8 +176,12 @@ async function runClaudeCode(ctx: AgentContext, model: string | null): Promise<A
 
   log("K7R2M4N8", `prompt for role=${ctx.role} (length=${fullPrompt.length}):\n${fullPrompt}`);
 
-  // Try resuming a cached session for re-entry scenarios (e.g. reviewer reject → developer re-entry).
+  // Try resuming a cached session.  This covers both normal re-entry
-  if (!ctx.isFirstVisit) {
+  // (e.g. reviewer reject → developer re-entry) AND the case where a
+  // previous run completed but frontmatter validation failed — the step
+  // was never written to CAS so isFirstVisit is still true, but the
+  // session cache holds a valid session we should resume.
+  {
     const cachedSessionId = await getCachedSessionId(
       "claude-code",
       ctx.threadId,

packages/agent-claude-code/src/cli.ts

@@ -1,4 +1,4 @@
-#!/usr/bin/env node
+#!/usr/bin/env -S node --disable-warning=ExperimentalWarning
 
 // eslint-disable-next-line -- dynamic import for version
 const pkg = await import("../package.json", { with: { type: "json" } });

packages/agent-hermes/__tests__/issue-551.test.ts

@@ -15,7 +15,8 @@ describe("Issue #551 — bin entry & engines", () => {
     const pkg = JSON.parse(readFileSync(join(PKG_ROOT, "package.json"), "utf-8"));
     const binPath = pkg.bin["uwf-hermes"];
     const content = readFileSync(join(PKG_ROOT, binPath), "utf-8");
-    expect(content.startsWith("#!/usr/bin/env node")).toBe(true);
+    expect(content.startsWith("#!/usr/bin/env")).toBe(true);
+    expect(content).toContain("node");
   });
 
   test("README.md explains uwf-hermes is an adapter", () => {

packages/agent-hermes/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@united-workforce/agent-hermes",
-  "version": "0.1.2",
+  "version": "0.1.4",
   "files": [
     "src",
     "dist",

packages/agent-hermes/src/acp-client.ts

@@ -12,7 +12,11 @@ const OWN_VERSION = (
   }
 ).version;
 
-const HERMES_COMMAND = "hermes";
+/** Resolve hermes binary: `UWF_HERMES_BIN` override → default `"hermes"` via PATH. */
+function resolveHermesCommand(): string {
+  const override = process.env.UWF_HERMES_BIN;
+  return override !== undefined && override !== "" ? override : "hermes";
+}
 const PROTOCOL_VERSION = 1;
 
 type JsonRpcResponse = {
@@ -271,7 +275,8 @@ export class HermesAcpClient {
       return;
     }
 
-    const child = spawn(HERMES_COMMAND, ["acp"], {
+    const hermesCommand = resolveHermesCommand();
+    const child = spawn(hermesCommand, ["acp"], {
       env: process.env,
       shell: false,
       stdio: ["pipe", "pipe", "pipe"],

packages/agent-hermes/src/cli.ts

@@ -1,4 +1,4 @@
-#!/usr/bin/env node
+#!/usr/bin/env -S node --disable-warning=ExperimentalWarning
 
 // eslint-disable-next-line -- dynamic import for version
 const pkg = await import("../package.json", { with: { type: "json" } });

packages/agent-hermes/src/hermes.ts

@@ -6,6 +6,7 @@ import {
   type AgentRunResult,
   buildContinuationPrompt,
   buildRolePrompt,
+  buildThreadProgress,
   createAgent,
 } from "@united-workforce/util-agent";
 import type { AcpUsage } from "./acp-client.js";
@@ -60,6 +61,9 @@ export function buildHermesPrompt(ctx: AgentContext): string {
     parts.push(ctx.outputFormatInstruction, "");
   }
 
+  // Inject thread progress so the agent knows step count and role visit count
+  parts.push(buildThreadProgress(ctx.steps, ctx.role), "");
+
   if (!ctx.isFirstVisit) {
     // Re-entry: show only steps since last visit, meta only
     parts.push(buildContinuationPrompt(ctx.steps, ctx.role, ctx.edgePrompt));
@@ -106,11 +110,15 @@ async function prepareSession(
   cwd: string,
   resumeDisabled: boolean,
 ): Promise<PromptAttempt> {
-  if (ctx.isFirstVisit || resumeDisabled) {
+  if (resumeDisabled) {
     await client.connect(cwd);
     return { useContinuation: false, resumed: false };
   }
 
+  // Check session cache regardless of isFirstVisit.  A previous run may
+  // have completed and cached its session but failed frontmatter
+  // validation — the step never got written to CAS so isFirstVisit is
+  // still true, yet we should resume the existing session.
   const cachedSessionId = await getCachedSessionId(ctx.threadId, ctx.role, ctx.storageRoot);
   if (cachedSessionId === null) {
     log("6RWK3N8Q", `no cached session for ${ctx.threadId}:${ctx.role}, starting new session`);

packages/agent-mock/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@united-workforce/agent-mock",
-  "version": "0.1.1",
+  "version": "0.1.2",
   "files": [
     "src",
     "dist",

packages/agent-mock/src/cli.ts

@@ -1,4 +1,4 @@
-#!/usr/bin/env node
+#!/usr/bin/env -S node --disable-warning=ExperimentalWarning
 
 // eslint-disable-next-line -- dynamic import for version
 const pkg = await import("../package.json", { with: { type: "json" } });

packages/cli/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@united-workforce/cli",
-  "version": "0.2.0",
+  "version": "0.3.0",
   "files": [
     "src",
     "dist",

packages/cli/src/__tests__/current-role.test.ts

@@ -28,9 +28,13 @@ roles:
       $status: "ready"
     frontmatter:
       type: object
-      required: ["$status"]
+      oneOf:
-      properties:
+        - properties:
-        $status: { type: string, enum: ["ready", "not-ready"] }
+            $status: { const: "ready" }
+          required: ["$status"]
+        - properties:
+            $status: { const: "not-ready" }
+          required: ["$status"]
   roleB:
     description: Second role
     goal: Do B
@@ -42,7 +46,7 @@ roles:
       type: object
       required: ["$status"]
       properties:
-        $status: { type: string, enum: ["done"] }
+        $status: { const: "done" }
 graph:
   $START:
     new:
@@ -82,9 +86,13 @@ roles:
       $status: "pass"
     frontmatter:
       type: object
-      required: ["$status"]
+      oneOf:
-      properties:
+        - properties:
-        $status: { type: string, enum: ["pass", "fail"] }
+            $status: { const: "pass" }
+          required: ["$status"]
+        - properties:
+            $status: { const: "fail" }
+          required: ["$status"]
   roleB:
     description: Pass role
     goal: Do B
@@ -96,7 +104,7 @@ roles:
       type: object
       required: ["$status"]
       properties:
-        $status: { type: string, enum: ["done"] }
+        $status: { const: "done" }
   roleC:
     description: Fail role
     goal: Do C
@@ -108,7 +116,7 @@ roles:
       type: object
       required: ["$status"]
       properties:
-        $status: { type: string, enum: ["done"] }
+        $status: { const: "done" }
 graph:
   $START:
     new:
@@ -155,7 +163,7 @@ roles:
       type: object
       required: ["$status"]
       properties:
-        $status: { type: string, enum: ["done"] }
+        $status: { const: "done" }
 graph:
   $START:
     new:

packages/cli/src/__tests__/prompt.test.ts

@@ -71,12 +71,22 @@ describe("prompt commands", () => {
   test("prompt bootstrap returns framework-agnostic setup instructions", () => {
     const result = cmdPromptBootstrap();
     expect(typeof result).toBe("string");
+    // Skills installation
     expect(result).toContain("uwf prompt usage");
     expect(result).toContain("uwf prompt workflow-authoring");
     expect(result).toContain("uwf prompt adapter-developing");
     expect(result).toContain("uwf-usage");
     expect(result).toContain("uwf-workflow-authoring");
     expect(result).toContain("uwf-adapter-developing");
+    // Fresh install scenario
+    expect(result).toContain("Fresh Install");
+    expect(result).toContain("uwf setup");
+    expect(result).toContain("--provider");
+    expect(result).toContain("--api-key");
+    expect(result).toContain("agent adapter");
+    // Upgrade scenario
+    expect(result).toContain("Upgrade");
+    expect(result).toContain("Migrate");
     // Should NOT contain Hermes-specific paths
     expect(result).not.toContain("~/.hermes/skills/");
     expect(result).not.toContain("> ~/.hermes/");

packages/cli/src/__tests__/solve-issue-tea-worktree.test.ts

@@ -21,11 +21,11 @@ describe("solve-issue workflow: Gitea API PR creation", () => {
     "..",
     "..",
     "..",
-    ".workflows",
+    "examples",
     "solve-issue.yaml",
   );
 
-  test("committer procedure should use curl API instead of tea pr create", async () => {
+  test("committer procedure should create PR via tea pr create", async () => {
     const yamlContent = await readFile(workflowPath, "utf-8");
     const workflow = parse(yamlContent) as WorkflowPayload;
 
@@ -33,25 +33,22 @@ describe("solve-issue workflow: Gitea API PR creation", () => {
     const committerProcedure = workflow.roles.committer?.procedure;
     expect(committerProcedure).toBeDefined();
 
-    // Verify the procedure uses curl API, not tea pr create
+    // Verify the procedure uses tea pr create for PR creation
-    expect(committerProcedure).toContain("curl");
+    expect(committerProcedure).toContain("tea pr create");
-    expect(committerProcedure).toContain("api/v1/repos");
+    expect(committerProcedure).toContain("git push");
-    expect(committerProcedure).toContain("/pulls");
+    expect(committerProcedure).toContain("Fixes #N");
-
-    // Verify it explicitly warns against tea pr create
-    expect(committerProcedure).toMatch(/do NOT use.*tea pr create/i);
   });
 
-  test("committer procedure should reference repoRemote from task prompt", async () => {
+  test("committer procedure should extract owner/repo from git remote", async () => {
     const yamlContent = await readFile(workflowPath, "utf-8");
     const workflow = parse(yamlContent) as WorkflowPayload;
 
     const committerProcedure = workflow.roles.committer?.procedure;
     expect(committerProcedure).toBeDefined();
 
-    // Verify the procedure mentions repoRemote is provided in task prompt
+    // Verify the procedure extracts owner/repo from remote
-    expect(committerProcedure).toMatch(/repo remote.*provided.*task prompt/i);
+    expect(committerProcedure).toContain("git remote get-url origin");
-    expect(committerProcedure).toMatch(/owner\/repo/i);
+    expect(committerProcedure).toContain("hook_failed");
   });
 
   test("committer procedure should include error handling for curl failures", async () => {
@@ -100,45 +97,42 @@ describe("solve-issue workflow: Gitea API PR creation", () => {
     expect(committedVariant.required).toContain("$status");
   });
 
-  test("developer procedure should include mandatory verification step", async () => {
+  test("developer procedure should include worktree setup", async () => {
     const yamlContent = await readFile(workflowPath, "utf-8");
     const workflow = parse(yamlContent) as WorkflowPayload;
 
     const developerProcedure = workflow.roles.developer?.procedure;
     expect(developerProcedure).toBeDefined();
 
-    // Verify the procedure includes mandatory verification step
+    // Verify the procedure includes worktree setup
-    expect(developerProcedure).toContain("MANDATORY VERIFICATION");
+    expect(developerProcedure).toContain("IMPORTANT");
-    expect(developerProcedure).toContain("git branch --show-current");
+    expect(developerProcedure).toContain("git worktree add");
-    expect(developerProcedure).toContain("git status");
+    expect(developerProcedure).toContain("pnpm install");
-    expect(developerProcedure).toMatch(/ls -la|verify.*exist/i);
   });
 
-  test("reviewer procedure should enforce worktree path verification", async () => {
+  test("reviewer procedure should verify branch and run checks", async () => {
     const yamlContent = await readFile(workflowPath, "utf-8");
     const workflow = parse(yamlContent) as WorkflowPayload;
 
     const reviewerProcedure = workflow.roles.reviewer?.procedure;
     expect(reviewerProcedure).toBeDefined();
 
-    // Verify the procedure includes critical enforcement
+    // Verify the procedure includes branch verification and build checks
-    expect(reviewerProcedure).toContain("CRITICAL");
+    expect(reviewerProcedure).toContain("git branch --show-current");
-    expect(reviewerProcedure).toMatch(/cd.*pwd/);
+    expect(reviewerProcedure).toContain("pnpm run build");
-    expect(reviewerProcedure).toContain(
+    expect(reviewerProcedure).toContain("pnpm run check");
-      "Do NOT report results without running the actual commands",
-    );
   });
 
-  test("developer procedure should include test debugging escalation", async () => {
+  test("developer procedure should include changeset and failure handling", async () => {
     const yamlContent = await readFile(workflowPath, "utf-8");
     const workflow = parse(yamlContent) as WorkflowPayload;
 
     const developerProcedure = workflow.roles.developer?.procedure;
     expect(developerProcedure).toBeDefined();
 
-    // Verify the procedure includes test failure guidance
+    // Verify the procedure includes changeset requirement and failure path
-    expect(developerProcedure).toMatch(/tests fail.*first run/i);
+    expect(developerProcedure).toContain(".changeset/");
-    expect(developerProcedure).toMatch(/3 test cycles|after 3 attempts/i);
     expect(developerProcedure).toContain("$status=failed");
+    expect(developerProcedure).toContain("pnpm test");
   });
 });

packages/cli/src/__tests__/thread-location.test.ts

@@ -54,7 +54,7 @@ roles:
       type: object
       required: ["$status"]
       properties:
-        $status: { type: string, enum: ["ready"] }
+        $status: { const: "ready" }
 graph:
   $START:
     new:
@@ -114,7 +114,7 @@ roles:
       type: object
       required: ["$status"]
       properties:
-        $status: { type: string, enum: ["ready"] }
+        $status: { const: "ready" }
 graph:
   $START:
     new:
@@ -161,7 +161,7 @@ roles:
       type: object
       required: ["$status"]
       properties:
-        $status: { type: string, enum: ["ready"] }
+        $status: { const: "ready" }
 graph:
   $START:
     new:

packages/cli/src/__tests__/thread-show-status.test.ts

@@ -31,7 +31,7 @@ roles:
       type: object
       required: ["$status"]
       properties:
-        $status: { type: string, enum: ["ready"] }
+        $status: { const: "ready" }
 graph:
   $START:
     new:

packages/cli/src/__tests__/thread-start-cwd-cli.test.ts

@@ -54,7 +54,7 @@ roles:
       type: object
       required: ["$status"]
       properties:
-        $status: { type: string, enum: ["ready"] }
+        $status: { const: "ready" }
 graph:
   $START:
     new:

packages/cli/src/__tests__/validate-semantic.test.ts

@@ -17,7 +17,7 @@ function makeWorkflow(overrides?: Partial<WorkflowPayload>): WorkflowPayload {
         frontmatter: {
           type: "object",
           properties: {
-            $status: { enum: ["done"] },
+            $status: { const: "done" },
             plan: { type: "string" },
           },
           required: ["$status", "plan"],
@@ -85,7 +85,7 @@ describe("Suite 1: Role Reference Integrity", () => {
       output: "None",
       frontmatter: {
         type: "object",
-        properties: { $status: { enum: ["done"] } },
+        properties: { $status: { const: "done" } },
         required: ["$status"],
       } as unknown as string,
     };
@@ -187,7 +187,7 @@ describe("Suite 2: Graph Structure", () => {
       output: "Isolated",
       frontmatter: {
         type: "object",
-        properties: { $status: { enum: ["done"] } },
+        properties: { $status: { const: "done" } },
         required: ["$status"],
       } as unknown as string,
     };
@@ -272,8 +272,8 @@ describe("Suite 3: Status-Edge Consistency", () => {
   });
 });
 
-describe("Suite 3b: Enum-Based Multi-Exit", () => {
+describe("Suite 3b: Enum-Based $status is Rejected", () => {
-  test("3b.1 enum multi-exit passes with matching graph keys", () => {
+  test("3b.1 enum multi-exit is rejected (must use oneOf + const)", () => {
     const wf = makeWorkflow();
     wf.roles.reviewer = {
       ...wf.roles.reviewer,
@@ -291,52 +291,10 @@ describe("Suite 3b: Enum-Based Multi-Exit", () => {
       rejected: { role: "writer", prompt: "Fix: {{{comments}}}", location: null },
     };
     const errors = validateWorkflow(wf);
-    expect(errors).toEqual([]);
+    expect(errors.some((e) => e.includes("must define") && e.includes("const"))).toBe(true);
   });
 
-  test("3b.2 enum multi-exit with extra graph key", () => {
+  test("3b.2 enum single-exit is rejected (must use const)", () => {
-    const wf = makeWorkflow();
-    wf.roles.reviewer = {
-      ...wf.roles.reviewer,
-      frontmatter: {
-        type: "object",
-        properties: {
-          $status: { enum: ["approved", "rejected"] },
-          comments: { type: "string" },
-        },
-        required: ["$status", "comments"],
-      } as unknown as string,
-    };
-    wf.graph.reviewer = {
-      approved: { role: "$END", prompt: "Done", location: null },
-      rejected: { role: "writer", prompt: "Fix", location: null },
-      timeout: { role: "$END", prompt: "Timed out", location: null },
-    };
-    const errors = validateWorkflow(wf);
-    expect(errors.some((e) => e.includes("extra status keys: timeout"))).toBe(true);
-  });
-
-  test("3b.3 enum multi-exit with missing graph key", () => {
-    const wf = makeWorkflow();
-    wf.roles.reviewer = {
-      ...wf.roles.reviewer,
-      frontmatter: {
-        type: "object",
-        properties: {
-          $status: { enum: ["approved", "rejected"] },
-          comments: { type: "string" },
-        },
-        required: ["$status", "comments"],
-      } as unknown as string,
-    };
-    wf.graph.reviewer = {
-      approved: { role: "$END", prompt: "Done", location: null },
-    };
-    const errors = validateWorkflow(wf);
-    expect(errors.some((e) => e.includes("missing status keys: rejected"))).toBe(true);
-  });
-
-  test("3b.4 enum with single explicit value passes", () => {
     const wf = makeWorkflow();
     wf.roles.writer = {
       ...wf.roles.writer,
@@ -351,28 +309,71 @@ describe("Suite 3b: Enum-Based Multi-Exit", () => {
     };
     wf.graph.writer = { ready: { role: "reviewer", prompt: "Review: {{{plan}}}", location: null } };
     const errors = validateWorkflow(wf);
-    expect(errors).toEqual([]);
+    expect(errors.some((e) => e.includes("must define") && e.includes("const"))).toBe(true);
   });
+});
 
-  test("3b.5 enum multi-exit mustache var not in frontmatter", () => {
+describe("Suite 3c: Const-Based Flat Schema", () => {
+  test("3c.1 flat schema with const $status passes validation", () => {
     const wf = makeWorkflow();
-    wf.roles.reviewer = {
+    wf.roles.writer = {
-      ...wf.roles.reviewer,
+      ...wf.roles.writer,
       frontmatter: {
         type: "object",
         properties: {
-          $status: { enum: ["approved", "rejected"] },
+          $status: { const: "done" },
-          comments: { type: "string" },
+          plan: { type: "string" },
         },
-        required: ["$status", "comments"],
+        required: ["$status", "plan"],
       } as unknown as string,
     };
-    wf.graph.reviewer = {
+    const errors = validateWorkflow(wf);
-      approved: { role: "$END", prompt: "Done: {{{nonexistent}}}", location: null },
+    expect(errors).toEqual([]);
-      rejected: { role: "writer", prompt: "Fix: {{{comments}}}", location: null },
+  });
+
+  test("3c.2 flat schema with const $status detects extra graph key", () => {
+    const wf = makeWorkflow();
+    wf.roles.writer = {
+      ...wf.roles.writer,
+      frontmatter: {
+        type: "object",
+        properties: {
+          $status: { const: "done" },
+          plan: { type: "string" },
+        },
+        required: ["$status", "plan"],
+      } as unknown as string,
+    };
+    wf.graph.writer = {
+      done: { role: "reviewer", prompt: "Review.", location: null },
+      extra: { role: "$END", prompt: "Nope.", location: null },
     };
     const errors = validateWorkflow(wf);
-    expect(errors.some((e) => e.includes("nonexistent") && e.includes("not found"))).toBe(true);
+    expect(errors.some((e) => e.includes("extra status keys") && e.includes("extra"))).toBe(true);
+  });
+
+  test("3c.3 flat schema with const $status validates mustache vars", () => {
+    const wf = makeWorkflow();
+    wf.roles.writer = {
+      ...wf.roles.writer,
+      frontmatter: {
+        type: "object",
+        properties: {
+          $status: { const: "done" },
+          plan: { type: "string" },
+        },
+        required: ["$status", "plan"],
+      } as unknown as string,
+    };
+    wf.graph.writer = {
+      done: { role: "reviewer", prompt: "Review: {{{nonexistent}}}", location: null },
+    };
+    const errors = validateWorkflow(wf);
+    expect(
+      errors.some(
+        (e) => e.includes('prompt variable "nonexistent"') && e.includes('role "writer"'),
+      ),
+    ).toBe(true);
   });
 });
 
@@ -480,7 +481,7 @@ describe("Suite 6: Multiple Errors Collection", () => {
       output: "None",
       frontmatter: {
         type: "object",
-        properties: { $status: { enum: ["done"] } },
+        properties: { $status: { const: "done" } },
         required: ["$status"],
       } as unknown as string,
     };

packages/cli/src/__tests__/workflow-resolution.test.ts

@@ -31,7 +31,7 @@ function makeMinimalPayload(name: string, description: string): WorkflowPayload
         frontmatter: {
           type: "object",
           properties: {
-            $status: { type: "string", enum: ["done"] },
+            $status: { const: "done" },
           },
           required: ["$status"],
         } as unknown as CasRef,

packages/cli/src/cli.ts

@@ -1,4 +1,4 @@
-#!/usr/bin/env node
+#!/usr/bin/env -S node --disable-warning=ExperimentalWarning
 
 import type { CasRef, ThreadId, ThreadStatus } from "@united-workforce/protocol";
 import { Command } from "commander";
@@ -11,7 +11,7 @@ import {
   cmdPromptUsage,
   cmdPromptWorkflowAuthoring,
 } from "./commands/prompt.js";
-import { cmdSetup, cmdSetupInteractive } from "./commands/setup.js";
+import { cmdSetup, cmdSetupInteractive, resolvePresetBaseUrl } from "./commands/setup.js";
 import { cmdStepFork, cmdStepList, cmdStepRead, cmdStepShow } from "./commands/step.js";
 import {
   cmdThreadCancel,
@@ -542,7 +542,7 @@ prompt
 
 program
   .command("setup")
-  .description("Configure provider, model, and agent")
+  .description("Configure provider, model, and agent. Run without options for interactive wizard.")
   .option("--provider <name>", "Provider name")
   .option("--base-url <url>", "OpenAI-compatible API base URL")
   .option("--api-key <key>", "API key")
@@ -558,10 +558,14 @@ program
     }) => {
       const storageRoot = resolveStorageRoot();
       runAction(async () => {
-        if (opts.provider && opts.baseUrl && opts.apiKey && opts.model) {
+        // Resolve preset base-url when provider is known but --base-url is omitted
+        const resolvedBaseUrl =
+          opts.baseUrl ??
+          (opts.provider !== undefined ? resolvePresetBaseUrl(opts.provider) : null);
+        if (opts.provider && resolvedBaseUrl && opts.apiKey && opts.model) {
           const result = await cmdSetup({
             provider: opts.provider,
-            baseUrl: opts.baseUrl,
+            baseUrl: resolvedBaseUrl,
             apiKey: opts.apiKey,
             model: opts.model,
             agent: opts.agent ?? undefined,
@@ -572,7 +576,7 @@ program
           await cmdSetupInteractive(storageRoot);
         } else {
           throw new Error(
-            "Non-interactive setup requires all of: --provider, --base-url, --api-key, --model",
+            "Non-interactive setup requires: --provider, --api-key, --model (--base-url is optional for preset providers)",
           );
         }
       });

packages/cli/src/commands/prompt.ts

@@ -1,10 +1,35 @@
+import { readFileSync } from "node:fs";
+import { dirname, join } from "node:path";
+import { fileURLToPath } from "node:url";
 import {
   generateAdapterDevelopingReference,
   generateUsageReference,
   generateWorkflowAuthoringReference,
-  VERSION,
 } from "@united-workforce/util";
 
+// CLI package version (for bootstrap prompt — uwf --version prints this)
+// Walk up from __dirname to find the nearest package.json (works from both src/ and dist/)
+function _findCliVersion(): string {
+  let dir = dirname(fileURLToPath(import.meta.url));
+  for (let i = 0; i < 5; i++) {
+    const candidate = join(dir, "package.json");
+    try {
+      const pkg = JSON.parse(readFileSync(candidate, "utf-8")) as {
+        name?: string;
+        version?: string;
+      };
+      if (pkg.name === "@united-workforce/cli") {
+        return pkg.version ?? "0.0.0";
+      }
+    } catch {
+      // not found, keep walking
+    }
+    dir = dirname(dir);
+  }
+  return "0.0.0";
+}
+const CLI_VERSION = _findCliVersion();
+
 export {
   generateAdapterDevelopingReference as cmdPromptAdapterDeveloping,
   generateUsageReference as cmdPromptUsage,
@@ -22,33 +47,275 @@ export function cmdPromptList(): ReadonlyArray<string> {
 }
 
 export function cmdPromptBootstrap(): string {
-  return `# uwf Skill Bootstrap
+  return `# uwf Bootstrap
 
-Install or update the uwf (United Workforce) skills for your agent.
+Set up or upgrade uwf (United Workforce) — from zero to running your first workflow.
 
-## Steps
+## Scenario A: Fresh Install
 
-1. **Check if uwf CLI is installed:**
+### Step 0 — Environment pre-flight check
+
+Run ALL checks below BEFORE installing anything. If any check fails, fix it first — do NOT skip ahead.
+
+\`\`\`bash
+# 1. Node.js
+node --version   # need v20+
+# FIX: install via nvm (https://github.com/nvm-sh/nvm) or your OS package manager
+
+# 2. Package manager — pnpm (preferred) or npm
+pnpm --version || npm --version
+# FIX for pnpm: corepack enable pnpm   (Node 20+)
+#   or: npm install -g pnpm
+# If pnpm is not an option, npm works too — adjust install commands below
+
+# 3. Global bin directory is in PATH
+# For pnpm:
+pnpm bin -g       # prints the global bin directory
+# For npm:
+npm prefix -g     # global prefix; bin is <prefix>/bin
+# The directory printed above MUST be in your PATH.
+# Test: echo \\$PATH | tr ':' '\\\\n'
+# FIX: add the missing directory to your shell's startup file
+#   (e.g. ~/.bashrc, ~/.zshrc, ~/.profile, or fish config):
+#   export PATH="<global-bin-dir>:$PATH"
+\`\`\`
+
+**All checks must pass before continuing.** If you had to modify PATH, verify the change persists by opening a new shell or sourcing your shell config.
+
+### Step 1 — Discover agents and install adapter
+
+**First, detect which supported agents are already installed on the user's machine:**
+
+\`\`\`bash
+# Check for Hermes Agent
+which hermes 2>/dev/null && hermes --version
+
+# Check for Claude Code
+which claude 2>/dev/null && claude --version   # should show "X.Y.Z (Claude Code)"
+\`\`\`
+
+**Based on the results:**
+
+- **Only hermes found** → install \`uwf-hermes\` adapter
+- **Only claude found** → install \`uwf-claude-code\` adapter
+- **Both found** → ask the user which agent they want uwf to use as default
+- **Neither found** → the user must install at least one agent first:
+  - Hermes Agent: https://hermes-agent.nousresearch.com/docs
+  - Claude Code: \`npm install -g @anthropic-ai/claude-code\`
+
+**Install the uwf CLI and the chosen adapter** using pnpm or npm:
+
+\`\`\`bash
+# CLI (required)
+pnpm add -g @united-workforce/cli       # or: npm install -g @united-workforce/cli
+
+# Adapter — install the one matching the detected agent:
+pnpm add -g @united-workforce/agent-hermes       # or: npm i -g @united-workforce/agent-hermes
+pnpm add -g @united-workforce/agent-claude-code   # or: npm i -g @united-workforce/agent-claude-code
+\`\`\`
+
+**⚠ Adapter versions are independent from CLI versions.** Do NOT try to match adapter version to CLI version. Just install \`@latest\` (the default).
+
+**After installing, verify that \`uwf\` and the adapter are available in PATH:**
+
+\`\`\`bash
+uwf --version          # should print ${CLI_VERSION}
+uwf-hermes --version   # or: uwf-claude-code --version
+\`\`\`
+
+If either command is not found, the global bin directory is not in the current shell's PATH. **You must fix this before continuing:**
+
+1. Find where the binary was installed:
    \`\`\`bash
-   uwf --help
+   find ~/.local ~/.hermes /usr/local -name uwf -type f 2>/dev/null
+   npm prefix -g    # global prefix — bin is <prefix>/bin
    \`\`\`
-   If not installed: \`npm install -g @united-workforce/cli\`
+2. Add the directory to PATH permanently by appending to the user's shell config (e.g. \`~/.bashrc\`, \`~/.zshrc\`, \`~/.profile\`, or fish config):
-   Current version: ${VERSION}
+   \`\`\`bash
+   export PATH="<global-bin-dir>:$PATH"
+   \`\`\`
+3. Source the updated config or open a new shell, then re-verify the commands work.
 
-2. **Install the skills** by running each command below and creating a skill from its output:
+**uwf-hermes** also requires the Hermes ACP plugin. Verify with \`hermes acp --help\`. If not available, install it:
-   - Run \`uwf prompt usage\` → create a skill named \`uwf-usage\`
+\`\`\`bash
-   - Run \`uwf prompt workflow-authoring\` → create a skill named \`uwf-workflow-authoring\`
+# Option A: install into hermes venv (recommended)
-   - Run \`uwf prompt adapter-developing\` → create a skill named \`uwf-adapter-developing\`
+source ~/.hermes/hermes-agent/.venv/bin/activate && pip install hermes-agent[acp]
 
-   Each command outputs a complete SKILL.md with YAML frontmatter — use your agent framework's skill creation API to save them.
+# Option B: pipx
+pipx install 'hermes-agent[acp]'
 
-3. **Verify** the skills are loadable by your agent framework.
+# Option C: if installed from source
+pip install -e '.[acp]'
+\`\`\`
 
-## Updating
+### Step 2 — Configure provider and model
 
-When \`uwf\` is upgraded, re-run \`uwf prompt bootstrap\` and follow the steps again.
+uwf needs an LLM provider to run agents. **Ask the user** for their provider, API key, and model, then run:
-The skill content is bundled with the CLI — always use \`uwf prompt <name>\` to get
+
-content matching your installed version.
+\`\`\`bash
+uwf setup --provider <name> --api-key <key> --model <model> --agent <adapter-command>
+\`\`\`
+
+**Note:** \`--agent\` takes the adapter **command name** (e.g. \`uwf-hermes\`), not the npm package name.
+
+**Preset providers** — when using a preset name, \`--base-url\` is auto-filled and can be omitted:
+
+| Provider | Name | Default base URL |
+|----------|------|-----------------|
+| OpenAI | \`openai\` | https://api.openai.com/v1 |
+| xAI | \`xai\` | https://api.x.ai/v1 |
+| OpenRouter | \`openrouter\` | https://openrouter.ai/api/v1 |
+| Venice | \`venice\` | https://api.venice.ai/api/v1 |
+| Dashscope | \`dashscope\` | https://dashscope.aliyuncs.com/compatible-mode/v1 |
+| DeepSeek | \`deepseek\` | https://api.deepseek.com/v1 |
+| SiliconFlow | \`siliconflow\` | https://api.siliconflow.cn/v1 |
+| VolcEngine | \`volcengine\` | https://ark.cn-beijing.volces.com/api/v3 |
+| Kimi (Moonshot) | \`kimi\` | https://api.moonshot.cn/v1 |
+| GLM (Zhipu AI) | \`glm\` | https://open.bigmodel.cn/api/paas/v4 |
+| StepFun | \`stepfun\` | https://api.stepfun.com/v1 |
+| MiniMax | \`minimax\` | https://api.minimax.io/v1 |
+| Ollama (local) | \`ollama\` | http://localhost:11434/v1 |
+
+For **non-preset providers**, you must specify \`--base-url\` manually.
+
+Example:
+\`\`\`bash
+uwf setup --provider openrouter --api-key sk-or-... --model anthropic/claude-sonnet-4 --agent uwf-hermes
+\`\`\`
+
+If the user doesn't know what to choose, suggest \`openrouter\` with \`anthropic/claude-sonnet-4\` as a sensible default.
+
+Config is saved to \`~/.uwf/config.yaml\`. Verify with \`cat ~/.uwf/config.yaml\`.
+
+### Step 3 — Install skills
+
+Run each command and create a skill from its output:
+
+\`\`\`bash
+uwf prompt usage               # → save as skill "uwf-usage"
+uwf prompt workflow-authoring  # → save as skill "uwf-workflow-authoring"
+uwf prompt adapter-developing  # → save as skill "uwf-adapter-developing"
+\`\`\`
+
+Each command outputs a complete SKILL.md with YAML frontmatter. Use your agent framework's skill creation API to save them (e.g. \`skill_manage(action='create', name='uwf-usage', content=<output>)\`).
+
+Verify skills are installed by listing them (e.g. \`skills_list()\`) and confirming all three appear.
+
+**⚠ After saving all skills, start a new session** so the agent loads the updated skill content. Skills saved in the current session are not active until the next session.
+
+### Step 4 — Verify end-to-end
+
+Create a minimal workflow file to test your setup:
+
+\`\`\`bash
+cat > /tmp/hello.yaml << 'YAML'
+name: hello
+description: Minimal smoke test
+roles:
+  greeter:
+    description: "Greet the user"
+    goal: "Respond with a friendly greeting"
+    capabilities: []
+    procedure: "Write a short greeting based on the prompt."
+    output: "A greeting message."
+    frontmatter:
+      type: object
+      properties:
+        $status: { const: done }
+        message: { type: string }
+      required: [$status, message]
+graph:
+  $START:
+    new: { role: greeter, prompt: "Say hello to the user." }
+    resume: { role: greeter, prompt: "Greet the user again." }
+  greeter:
+    done: { role: "$END", prompt: "Done." }
+YAML
+\`\`\`
+
+Then run:
+
+\`\`\`bash
+uwf thread start /tmp/hello.yaml -p "Hello, world!"
+uwf thread exec <thread-id>
+uwf thread show <thread-id>
+\`\`\`
+
+If the thread reaches \`$END\` with status \`completed\`, the setup is working.
+
+## Scenario B: Upgrade from Previous Version
+
+### Step 1 — Update packages
+
+\`\`\`bash
+# Using pnpm
+pnpm add -g @united-workforce/cli@latest
+
+# Using npm
+npm install -g @united-workforce/cli@latest
+\`\`\`
+
+\`\`\`bash
+uwf --version   # should print ${CLI_VERSION}
+\`\`\`
+
+Also update your adapter(s):
+
+\`\`\`bash
+# pnpm
+pnpm add -g @united-workforce/agent-hermes@latest
+
+# npm
+npm install -g @united-workforce/agent-hermes@latest
+\`\`\`
+
+### Step 2 — Regenerate skills
+
+Skill content is bundled with the CLI — always regenerate after upgrading:
+
+\`\`\`bash
+uwf prompt usage               # → update skill "uwf-usage"
+uwf prompt workflow-authoring  # → update skill "uwf-workflow-authoring"
+uwf prompt adapter-developing  # → update skill "uwf-adapter-developing"
+\`\`\`
+
+**⚠ After updating skills, start a new session** to load the new skill content.
+
+### Step 3 — Migrate workflow YAML files (if needed)
+
+Check the changelog for breaking changes. Known migrations:
+
+- **v0.2.0**: \`$START._\` → \`$START.new\` + \`$START.resume\`. All workflow YAML files must be updated:
+  \`\`\`yaml
+  # Before (v0.1.x)
+  $START:
+    _: { role: planner, prompt: "..." }
+
+  # After (v0.2.0+)
+  $START:
+    new: { role: planner, prompt: "..." }
+    resume: { role: planner, prompt: "Review previous run and continue." }
+  \`\`\`
+
+Update all \`.workflow/\` and \`.workflows/\` YAML files in your projects. \`uwf workflow add\` will reject files with the old \`_\` syntax.
+
+- **v0.2.1**: \`$status: { enum: [value] }\` → \`$status: { const: "value" }\`. The validator no longer accepts \`enum\` for \`$status\`. Update all workflow YAML files:
+  \`\`\`yaml
+  # Before (v0.2.0)
+  $status: { enum: [done] }
+  $status: { type: string, enum: ["ready", "failed"] }
+
+  # After (v0.2.1+)
+  $status: { const: "done" }
+  # For multi-exit, use oneOf with const (unchanged)
+  \`\`\`
+
+### Step 4 — Verify
+
+\`\`\`bash
+uwf thread start <your-workflow> -p "upgrade test"
+uwf thread exec <thread-id>
+\`\`\`
 
 ## Available prompts
 
@@ -57,6 +324,7 @@ uwf prompt list                # list available prompt names
 uwf prompt usage               # CLI usage guide
 uwf prompt workflow-authoring  # workflow YAML design guide
 uwf prompt adapter-developing  # building agent adapters
+uwf prompt bootstrap           # this guide
 \`\`\`
 `;
 }

packages/cli/src/commands/setup.ts

@@ -1,3 +1,4 @@
+import { execFileSync } from "node:child_process";
 import { existsSync, mkdirSync, readdirSync, readFileSync, statSync, writeFileSync } from "node:fs";
 import { join } from "node:path";
 import { stdin as input, stdout as output } from "node:process";
@@ -72,6 +73,12 @@ const PRESET_PROVIDERS = [
   { name: "ollama", label: "Ollama (local)", baseUrl: "http://localhost:11434/v1" },
 ] as const;
 
+/** Look up the base URL for a preset provider name. Returns null if not a preset. */
+export function resolvePresetBaseUrl(providerName: string): string | null {
+  const preset = PRESET_PROVIDERS.find((p) => p.name === providerName);
+  return preset !== undefined ? preset.baseUrl : null;
+}
+
 type SetupArgs = {
   provider: string;
   baseUrl: string;
@@ -175,7 +182,6 @@ export async function _discoverAgents(): Promise<string[]> {
 
 async function _tryWhichDiscovery(): Promise<string[] | null> {
   try {
-    const { execFileSync } = await import("node:child_process");
     const text = execFileSync("which", ["-a", "uwf-hermes", "uwf-claude-code", "uwf-cursor"], {
       encoding: "utf-8",
       stdio: ["pipe", "pipe", "pipe"],
@@ -391,6 +397,37 @@ function mergeConfig(existing: Record<string, unknown>, args: SetupArgs): Record
   };
 }
 
+/**
+ * Check if the configured adapter binary (and its dependencies) are in PATH.
+ * Returns warnings array — empty means all good.
+ */
+export function _checkAdapterAvailability(agentName: string): string[] {
+  const warnings: string[] = [];
+  const binary = `uwf-${agentName}`;
+
+  try {
+    execFileSync("which", [binary], { encoding: "utf8", stdio: ["pipe", "pipe", "pipe"] });
+  } catch {
+    warnings.push(
+      `${binary} not found in PATH. Install it: pnpm add -g @united-workforce/agent-${agentName}`,
+    );
+    return warnings; // skip dependency check if adapter itself is missing
+  }
+
+  // uwf-hermes depends on hermes CLI
+  if (agentName === "hermes") {
+    try {
+      execFileSync("which", ["hermes"], { encoding: "utf8", stdio: ["pipe", "pipe", "pipe"] });
+    } catch {
+      warnings.push(
+        'hermes CLI not found in PATH (required by uwf-hermes). Fix: export PATH="$HOME/.hermes/hermes-agent/.venv/bin:$PATH"',
+      );
+    }
+  }
+
+  return warnings;
+}
+
 /**
  * Non-interactive setup. All required args provided via CLI flags.
  */
@@ -405,15 +442,26 @@ export async function cmdSetup(args: SetupArgs): Promise<Record<string, unknown>
 
   writeFileSync(configPath, stringify(merged, { indent: 2 }), "utf8");
 
+  // Print config path to stderr (stdout is reserved for JSON output)
+  console.error(`Config saved to ${configPath} ✓`);
+
   // Validate model connectivity
   const validation = await validateModel(args.baseUrl, args.apiKey, args.model);
 
+  // Check adapter availability
+  const agentName = _agentNameFromBinary(args.agent ?? "hermes");
+  const adapterWarnings = _checkAdapterAvailability(agentName);
+  for (const w of adapterWarnings) {
+    console.error(`⚠ ${w}`);
+  }
+
   return {
     configPath,
     provider: args.provider,
     model: args.model,
     defaultAgent: merged.defaultAgent,
     validation,
+    adapterWarnings,
   };
 }
 

packages/cli/src/commands/thread.ts

@@ -1004,6 +1004,12 @@ function spawnAgent(
     });
   } catch (e) {
     const err = e as NodeJS.ErrnoException & { stderr?: Buffer | string | null };
+    if (err.code === "ENOENT") {
+      failStep(
+        plog,
+        `"${agent.command}" not found in PATH. Install it or check your PATH config. Run: which ${agent.command}`,
+      );
+    }
     const stderr =
       err.stderr == null
         ? ""

packages/cli/src/validate-semantic.ts

@@ -24,22 +24,22 @@ function isOneOfSchema(fm: unknown): fm is SchemaObj & { oneOf: SchemaObj[] } {
   return Array.isArray(obj.oneOf);
 }
 
-/** Check if a frontmatter schema declares "$status" as an enum (the required form for user roles). */
+/** Check if a frontmatter schema declares "$status" as const (flat schema form). */
-function hasStatusEnum(fm: unknown): boolean {
+function hasStatusConst(fm: unknown): boolean {
   if (typeof fm !== "object" || fm === null) return false;
   const obj = fm as SchemaObj;
   const props = obj.properties as Record<string, SchemaObj> | undefined;
   if (!props?.$status) return false;
-  return Array.isArray(props.$status.enum);
+  return typeof props.$status.const === "string";
 }
 
-/** Extract status values from an enum-based $status field. */
+/** Extract status values from a const-based $status field. */
-function getEnumStatuses(fm: SchemaObj): string[] {
+function getConstStatuses(fm: SchemaObj): string[] {
   const props = fm.properties as Record<string, SchemaObj> | undefined;
   if (!props?.$status) return [];
   const statusDef = props.$status;
-  if (!Array.isArray(statusDef.enum)) return [];
+  if (typeof statusDef.const === "string") return [statusDef.const];
-  return statusDef.enum as string[];
+  return [];
 }
 
 /** Get property names from a schema object. */
@@ -248,21 +248,21 @@ function checkRoleConsistency(payload: WorkflowPayload, errors: string[]): void
       checkOneOfDiscriminant(roleName, variants, statuses, errors);
       checkStatusEdges(roleName, graphKeys, new Set(statuses), errors);
       checkMultiExitMustache(roleName, graphEntry, variants, errors);
-    } else if (hasStatusEnum(fm)) {
+    } else if (hasStatusConst(fm)) {
-      const statuses = getEnumStatuses(fm as SchemaObj);
+      const statuses = getConstStatuses(fm as SchemaObj);
       checkStatusEdges(roleName, graphKeys, new Set(statuses), errors);
-      // For enum-based schemas, mustache vars come from the flat properties
+      // For const-based flat schemas, mustache vars come from the flat properties
-      checkEnumMustache(roleName, graphEntry, fm as SchemaObj, errors);
+      checkFlatMustache(roleName, graphEntry, fm as SchemaObj, errors);
     } else {
       errors.push(
-        `role "${roleName}" must define "$status" as an enum (or oneOf const) in frontmatter`,
+        `role "${roleName}" must define "$status" as const (or oneOf with const) in frontmatter`,
       );
     }
   }
 }
 
 /** Check mustache vars in all edge prompts against flat schema properties. */
-function checkEnumMustache(
+function checkFlatMustache(
   roleName: string,
   graphEntry: Record<string, { role: string; prompt: string }>,
   fm: SchemaObj,

packages/cli/src/validate.ts

@@ -99,7 +99,7 @@ export function checkWorkflowFilenameConsistency(
 ): string | null {
   const expected = workflowNameFromPath(filePath);
   if (payload.name !== expected) {
-    return `workflow name mismatch: file "${basename(filePath)}" implies name "${expected}" but YAML declares name "${payload.name}"`;
+    return `workflow name mismatch: file "${basename(filePath)}" implies name "${expected}" but YAML declares name "${payload.name}". Either rename the file to "${payload.name}.yaml" or change the YAML \`name\` field to "${expected}"`;
   }
   return null;
 }

packages/eval/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@united-workforce/eval",
-  "version": "0.1.3",
+  "version": "0.1.5",
   "private": false,
   "files": [
     "src",

packages/eval/src/commands/list.ts

@@ -6,7 +6,7 @@ import { formatList, selectEntries } from "./format.js";
 import { readEvalEntries } from "./read.js";
 
 const log = createLogger({ sink: { kind: "stderr" } });
-const LOG_LIST = "L5KX9R2B";
+const LOG_LIST = "H5KX9R2B";
 
 type ListCliOptions = {
   task: string | undefined;

packages/util-agent/__tests__/build-output-format-instruction.test.ts

@@ -143,7 +143,7 @@ describe("buildOutputFormatInstruction", () => {
         {
           type: "object",
           properties: {
-            $status: { type: "string", enum: ["approved"] },
+            $status: { const: "approved" },
             branch: { type: "string" },
           },
           required: ["$status"],
@@ -151,7 +151,7 @@ describe("buildOutputFormatInstruction", () => {
         {
           type: "object",
           properties: {
-            $status: { type: "string", enum: ["rejected"] },
+            $status: { const: "rejected" },
             comments: { type: "string" },
           },
           required: ["$status"],
@@ -225,4 +225,34 @@ describe("buildOutputFormatInstruction", () => {
     const result = buildOutputFormatInstruction({});
     expect(result).toContain("Focus exclusively on YOUR role");
   });
+
+  test("renders const value as literal in flat schema example", () => {
+    const schema = {
+      type: "object",
+      properties: {
+        $status: { type: "string", const: "greeted" },
+        message: { type: "string" },
+      },
+      required: ["$status", "message"],
+    };
+    const result = buildOutputFormatInstruction(schema);
+    expect(result).toContain("$status: greeted");
+    expect(result).toContain("fixed value");
+    expect(result).not.toContain("$status: <string>");
+  });
+
+  test("renders const value for non-string types", () => {
+    const schema = {
+      type: "object",
+      properties: {
+        count: { type: "number", const: 42 },
+        done: { type: "boolean", const: true },
+      },
+      required: ["count", "done"],
+    };
+    const result = buildOutputFormatInstruction(schema);
+    expect(result).toContain("count: 42");
+    expect(result).toContain("done: true");
+    expect(result).toContain("fixed value");
+  });
 });

packages/util-agent/__tests__/build-thread-progress.test.ts

@@ -0,0 +1,59 @@
+import type { StepContext } from "@united-workforce/protocol";
+import { describe, expect, test } from "vitest";
+import { buildThreadProgress } from "../src/build-thread-progress.js";
+
+function makeStep(role: string): StepContext {
+  return {
+    role,
+    output: {},
+    detail: "0000000000000" as string,
+    agent: "uwf-mock",
+    edgePrompt: "",
+    startedAtMs: 0,
+    completedAtMs: 0,
+    cwd: "",
+    assembledPrompt: null,
+    usage: null,
+    content: null,
+  };
+}
+
+describe("buildThreadProgress", () => {
+  test("first step of thread", () => {
+    const result = buildThreadProgress([], "proponent");
+    expect(result).toContain("## Thread Progress");
+    expect(result).toContain("first step");
+    expect(result).toContain("first time");
+    expect(result).toContain("proponent");
+  });
+
+  test("second step, role not seen before", () => {
+    const steps = [makeStep("opponent")];
+    const result = buildThreadProgress(steps, "proponent");
+    expect(result).toContain("Thread step 2");
+    expect(result).toContain("spoken 0 times");
+  });
+
+  test("role has spoken once before", () => {
+    const steps = [makeStep("proponent"), makeStep("opponent")];
+    const result = buildThreadProgress(steps, "proponent");
+    expect(result).toContain("Thread step 3");
+    expect(result).toContain("spoken 1 time before");
+    // singular "time" not "times"
+    expect(result).not.toContain("1 times");
+  });
+
+  test("role has spoken multiple times", () => {
+    const steps = [
+      makeStep("proponent"),
+      makeStep("opponent"),
+      makeStep("proponent"),
+      makeStep("opponent"),
+      makeStep("proponent"),
+      makeStep("opponent"),
+    ];
+    const result = buildThreadProgress(steps, "proponent");
+    expect(result).toContain("Thread step 7");
+    expect(result).toContain("spoken 3 times");
+  });
+});

packages/util-agent/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@united-workforce/util-agent",
-  "version": "0.1.0",
+  "version": "0.1.1",
   "files": [
     "src",
     "dist",

packages/util-agent/src/build-output-format-instruction.ts

@@ -74,6 +74,10 @@ function collectObjectSchemas(schema: JSONSchema): JSONSchema[] {
 }
 
 function resolvePropertySchema(prop: JSONSchema): JSONSchema {
+  if (prop.const !== undefined) {
+    return prop;
+  }
+
   if (Array.isArray(prop.enum) && prop.enum.length > 0) {
     return prop;
   }
@@ -113,6 +117,11 @@ function buildPropertyExampleLine(prop: SchemaProperty): string {
     commentParts.push("required");
   }
 
+  if (resolved.const !== undefined) {
+    commentParts.push("fixed value");
+    return `${prop.name}: ${formatYamlScalar(resolved.const)}${buildPropertyComment(commentParts)}`;
+  }
+
   if (Array.isArray(resolved.enum) && resolved.enum.length > 0) {
     const enumValues = resolved.enum.map((v) => String(v));
     commentParts.push(...enumValues);

packages/util-agent/src/build-thread-progress.ts

@@ -0,0 +1,27 @@
+import type { StepContext } from "@united-workforce/protocol";
+
+/**
+ * Build a compact thread-progress summary so the agent knows where it is
+ * in the conversation without making tool calls to count steps.
+ *
+ * Example output:
+ *   ## Thread Progress
+ *   Thread step 6. You (proponent) have spoken 2 times before this turn.
+ */
+export function buildThreadProgress(steps: StepContext[], role: string): string {
+  const totalSteps = steps.length;
+  const roleVisits = steps.filter((s) => s.role === role).length;
+
+  const parts = [`## Thread Progress`];
+  if (totalSteps === 0) {
+    parts.push(
+      `This is the first step of the thread. You (${role}) are speaking for the first time.`,
+    );
+  } else {
+    parts.push(
+      `Thread step ${totalSteps + 1}. You (${role}) have spoken ${roleVisits} time${roleVisits === 1 ? "" : "s"} before this turn.`,
+    );
+  }
+
+  return parts.join("\n");
+}

packages/util-agent/src/index.ts

@@ -1,6 +1,7 @@
 export { buildContinuationPrompt } from "./build-continuation-prompt.js";
 export { buildOutputFormatInstruction } from "./build-output-format-instruction.js";
 export { buildRolePrompt } from "./build-role-prompt.js";
+export { buildThreadProgress } from "./build-thread-progress.js";
 export type { BuildContextMeta } from "./context.js";
 export { buildContext, buildContextWithMeta } from "./context.js";
 export type { ExtractResult, ResolvedLlmProvider } from "./extract.js";

packages/util/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@united-workforce/util",
-  "version": "0.1.2",
+  "version": "0.1.4",
   "files": [
     "src",
     "dist",

packages/util/src/usage-reference.ts

@@ -140,5 +140,18 @@ For specific scenarios, run the corresponding \`uwf prompt\` command:
 |----------|---------|-------------|
 | Writing workflow YAML | \`uwf prompt workflow-authoring\` | Designing roles, conditions, graphs, and edge prompts |
 | Building a new agent adapter | \`uwf prompt adapter-developing\` | Creating a new \`uwf-<name>\` CLI adapter |
+
+## Upgrading
+
+\`\`\`bash
+# Install the latest version
+pnpm add -g @united-workforce/cli@latest @united-workforce/agent-hermes@latest
+# or: npm install -g @united-workforce/cli@latest @united-workforce/agent-hermes@latest
+
+# Verify
+uwf --version
+
+# Then run uwf prompt bootstrap and follow the upgrade instructions
+\`\`\`
 `;
 }

packages/util/src/workflow-authoring-reference.ts

@@ -28,6 +28,7 @@ roles:                         # named actors
       2. Do that
     output: "..."              # what the agent should produce
     frontmatter:               # JSON Schema for structured output
+      type: object
       oneOf:
         - properties:
             $status: { const: "ready" }
@@ -71,10 +72,13 @@ The \`frontmatter\` field is a standard JSON Schema. It defines the structured f
 
 ### \`$status\` Field
 
-\`$status\` is the only standard field. Its value determines which graph edge the moderator follows. Use \`const\` to constrain each variant:
+\`$status\` is the only standard field. Its value determines which graph edge the moderator follows.
+
+**Multi-exit (oneOf)** — use \`const\` to constrain each variant:
 
 \`\`\`yaml
 frontmatter:
+  type: object
   oneOf:
     - properties:
         $status: { const: "done" }
@@ -86,22 +90,26 @@ frontmatter:
       required: [$status, error]
 \`\`\`
 
-### Custom Fields
+**Single-exit (flat schema)** — same syntax, just no \`oneOf\` wrapper:
-
-Add any fields you need for data passing between roles. These are available in edge prompts via Mustache templates.
-
-### Flat Schema (Single Status)
-
-When a role has only one outcome:
 
 \`\`\`yaml
 frontmatter:
+  type: object
   properties:
     $status: { const: "done" }
     summary: { type: string }
   required: [$status, summary]
 \`\`\`
 
+**Important rules:**
+- \`type: object\` is **required** at the top level of frontmatter (both flat and oneOf)
+- \`$status\` always uses \`const: "value"\` — simple and consistent
+- \`enum\` is **not supported** for \`$status\` — the validator will reject it
+
+### Custom Fields
+
+Add any fields you need for data passing between roles. These are available in edge prompts via Mustache templates.
+
 ## Graph Routing
 
 The graph maps each role's \`$status\` values to the next role:

workflows/solve-issue.yaml

@@ -1,329 +0,0 @@
-name: solve-issue
-description: TDD-driven issue resolution adapted for the workflow monorepo with bun + vitest
-roles:
-  planner:
-    description: Analyzes issue and outputs a TDD test spec
-    goal: You are a planning agent. You analyze Gitea issues and produce a TDD test specification that downstream roles will implement and verify.
-    capabilities:
-    - issue-analysis
-    - planning
-    procedure: 'On first run (no previous steps):
-
-      1. Read the issue and all comments from Gitea using `tea issues <number> -r <owner/repo>`
-
-      2. Look for project conventions files (CLAUDE.md, CONTRIBUTING.md) in the repo
-
-      3. Assess whether the issue has enough information to produce a test spec
-
-      4. If insufficient info: comment on the issue via `echo "..." | tea comment <number> -r <owner/repo>` (skip if you already commented), then output $status=insufficient_info
-
-      5. If sufficient: produce a detailed TDD test spec in markdown covering all scenarios
-
-
-      On subsequent runs (bounced back by tester with fix_spec):
-
-      1. Read the tester''s output from the previous step to understand what''s wrong with the spec
-
-      2. Revise the test spec accordingly
-
-
-      After producing the test spec:
-
-      1. The test spec is stored in CAS automatically by the uwf pipeline (agents do not need to call `ocas put` directly)
-
-      2. Put the hash in frontmatter.plan (required when $status=ready)
-
-      3. Set repoPath to the absolute path of the repository root
-
-
-
-      IMPORTANT: Extract the repo remote (owner/repo) from git:
-
-      ```bash
-
-      git remote get-url origin | sed ''s|.*[:/]\([^/]*/[^.]*\).*|\1|''
-
-      ```
-
-      Store the result as repoRemote in your frontmatter output so downstream roles can use it for tea/API calls.'
-    output: Output a brief summary of the test spec. Set $status to ready (with plan hash and repoPath) or insufficient_info.
-    frontmatter:
-      oneOf:
-      - properties:
-          $status:
-            const: ready
-          plan:
-            type: string
-          repoPath:
-            type: string
-          repoRemote:
-            type: string
-        required:
-        - $status
-        - plan
-        - repoPath
-      - properties:
-          $status:
-            const: insufficient_info
-          reason:
-            type: string
-        required:
-        - $status
-        - reason
-  developer:
-    description: TDD implementation per test spec
-    goal: You are a developer agent. You implement code changes following TDD — write tests first, then implementation.
-    capabilities:
-    - coding
-    procedure: "IMPORTANT: Always work in a git worktree, NEVER modify the main working directory directly.\nThe repo path and other details are provided in your task prompt.\n\nBefore starting any work,\
-      \ set up an isolated worktree:\n1. cd into the repo path provided in your task prompt\n2. `git fetch origin` to get latest refs\n3. First time (no existing branch):\n   - `git worktree add .worktrees/fix/<issue-number>-<short-slug>\
-      \ -b fix/<issue-number>-<short-slug> origin/main`\n   - `cd .worktrees/fix/<issue-number>-<short-slug> && bun install`\n4. If bounced back from reviewer or tester (branch already exists):\n   - cd\
-      \ into the existing worktree under `.worktrees/fix/<issue-number>-<short-slug>`\n   - `git fetch origin && git rebase origin/main`\n5. ALL subsequent work must happen inside the worktree directory.\n\
-      \nThen implement TDD:\n6. Read the test spec from CAS: `ocas get <plan hash>` (find the hash from the planner's output in your task prompt)\n7. If bounced back from reviewer or tester: read the\
-      \ previous role's feedback in your task prompt\n8. Write tests first based on the spec (use vitest)\n9. Implement the code to make tests pass\n10. Ensure `bun run build` passes with no errors\n11.\
-      \ Run `bun test` to verify all tests pass\n\nIf you cannot complete the implementation (e.g. the issue is too complex, blocked by external factors,\nor repeated attempts fail), set $status=failed\
-      \ with a reason.\n"
-    output: List all files changed and provide a summary. Set $status to done (with branch/worktree), or failed (with reason).
-    frontmatter:
-      oneOf:
-      - properties:
-          $status:
-            const: done
-          branch:
-            type: string
-          worktree:
-            type: string
-          repoRemote:
-            type: string
-        required:
-        - $status
-        - branch
-        - worktree
-      - properties:
-          $status:
-            const: failed
-          reason:
-            type: string
-          repoRemote:
-            type: string
-        required:
-        - $status
-        - reason
-  reviewer:
-    description: Code standards compliance check
-    goal: You are a code reviewer. You verify code standards compliance — NOT functionality (that's the tester's job).
-    capabilities:
-    - code-review
-    - static-analysis
-    procedure: 'The worktree path is provided in your task prompt. cd into it first.
-
-
-      Before reviewing, verify the git branch:
-
-      1. Run `git branch --show-current` — confirm the branch name references the issue number being worked on
-
-      2. If the branch doesn''t correspond to the issue, flag it in your output and reject
-
-
-      Then perform code review:
-
-      Hard checks (must all pass):
-
-      3. `bun run build` — no build errors
-
-      4. `bunx biome check` — no lint violations
-
-      5. TypeScript strict mode — no type errors
-
-
-      Soft checks (review against project conventions from CLAUDE.md):
-
-      - Functional-first: functions + types, no classes (except for errors or third-party requirements)
-
-      - Named exports only, no default exports
-
-      - No optional properties (use `T | null` instead of `?:`)
-
-      - Folder module discipline: index.ts only re-exports, types in types.ts
-
-      - Crockford Base32 log tags (8-char, unique per call site)
-
-      - No `console.log` in production code (use createLogger from @united-workforce/util)
-
-      - No dynamic imports in production code
-
-
-      Only review standards compliance. Do NOT test functionality.
-
-      If rejecting, you MUST explain the specific reason in your output.
-
-      '
-    output: Explain your decision with specific file/line references. Set $status to approved (with branch/worktree) or rejected (with comments).
-    frontmatter:
-      oneOf:
-      - properties:
-          $status:
-            const: approved
-          branch:
-            type: string
-          worktree:
-            type: string
-          repoRemote:
-            type: string
-        required:
-        - $status
-        - branch
-        - worktree
-      - properties:
-          $status:
-            const: rejected
-          comments:
-            type: string
-          worktree:
-            type: string
-          repoRemote:
-            type: string
-        required:
-        - $status
-        - comments
-        - worktree
-  tester:
-    description: Functional correctness verification
-    goal: You are a tester agent. You verify that the implementation correctly satisfies every scenario in the test spec.
-    capabilities:
-    - testing
-    procedure: "The worktree path is provided in your task prompt. cd into it first.\n\n1. Run `bun test` for automated test verification\n2. Read the test spec from CAS: `ocas get <plan hash>` (find\
-      \ the hash from the planner step in the thread history)\n3. Verify each scenario in the spec is covered and passing\n4. Determine outcome:\n   - passed: all scenarios verified, tests pass\n   - fix_code:\
-      \ tests fail or implementation doesn't match spec → send back to developer\n   - fix_spec: the spec itself is wrong or incomplete → send back to planner\n"
-    output: Report test results per scenario. Set $status to passed (with branch/worktree), fix_code (with report), or fix_spec (with report).
-    frontmatter:
-      oneOf:
-      - properties:
-          $status:
-            const: passed
-          branch:
-            type: string
-          worktree:
-            type: string
-          repoRemote:
-            type: string
-        required:
-        - $status
-        - branch
-        - worktree
-      - properties:
-          $status:
-            const: fix_code
-          report:
-            type: string
-          repoRemote:
-            type: string
-          worktree:
-            type: string
-          branch:
-            type: string
-        required:
-        - $status
-        - report
-      - properties:
-          $status:
-            const: fix_spec
-          report:
-            type: string
-          repoRemote:
-            type: string
-          worktree:
-            type: string
-          branch:
-            type: string
-        required:
-        - $status
-        - report
-  committer:
-    description: Commits and creates PR
-    goal: You are a committer agent. You create a clean commit and push a PR linking the original issue.
-    capabilities: []
-    procedure: "The worktree path, branch name, and repo remote (owner/repo) are provided in your task prompt.\ncd into the worktree first.\n\nNote: You inherit the developer's worktree and branch. Do NOT\
-      \ create a new branch.\n1. Stage all changes: `git add -A`\n2. Commit with a descriptive message referencing the issue: `git commit -m \"type: description\\n\\nFixes #N\"`\n3. Push the branch: `git\
-      \ push -u origin <branch-name>`\n4. **Verify push succeeded** — run `git ls-remote origin <branch-name>` and confirm it prints a commit hash.\n   - If no output or push failed: capture the error, mark hook_failed\n\
-      5. Create a PR using the Gitea API (do NOT use `tea pr create` — it fails in worktrees):\n   ```bash\n   GITEA_TOKEN=$(cfg get GITEA_TOKEN)\n   curl -s -X POST -H \"Authorization: token $GITEA_TOKEN\" -H \"Content-Type: application/json\" \\\n\
-      \     \"https://git.shazhou.work/api/v1/repos/<owner>/<repo>/pulls\" \\\n     -d '{\"title\":\"...\",\"body\":\"...\",\"head\":\"<branch>\",\"base\":\"main\"}'\n   ```\n   - The repo remote (owner/repo format, e.g. \"shazhou/united-workforce\") is given in your task prompt — use it directly.\n\
-      \   - PR body must include: What / Why / Changes / Ref sections, with `Fixes #N` in Ref\n6. **Verify PR was created** — parse the curl response JSON: it must contain a `\"number\"` field. Print the PR URL.\n\
-      \   - If curl returns an error or no number field: capture the response, mark hook_failed\n7. After PR creation, clean up the worktree:\n   - cd to the repo root (parent of .worktrees)\n   - `git worktree remove <worktree-path>`"
-    output: Include PR URL on success or error log on failure. Set $status to committed (with prUrl) or hook_failed (with error).
-    frontmatter:
-      oneOf:
-      - properties:
-          $status:
-            const: committed
-          prUrl:
-            type: string
-          repoRemote:
-            type: string
-          worktree:
-            type: string
-          branch:
-            type: string
-        required:
-        - $status
-        - prUrl
-      - properties:
-          $status:
-            const: hook_failed
-          error:
-            type: string
-          repoRemote:
-            type: string
-          worktree:
-            type: string
-          branch:
-            type: string
-        required:
-        - $status
-        - error
-graph:
-  $START:
-    new:
-      role: planner
-      prompt: Analyze the issue and produce an implementation plan.
-    resume:
-      role: planner
-      prompt: Review the previous run output and continue the work.
-  planner:
-    insufficient_info:
-      role: $SUSPEND
-      prompt: "信息不足，需要补充：{{{reason}}}"
-    ready:
-      role: developer
-      prompt: 'Implement the TDD test spec (CAS hash: {{{plan}}}) in repo {{{repoPath}}}. Repo remote: {{{repoRemote}}}.'
-  developer:
-    done:
-      role: reviewer
-      prompt: 'Review branch {{{branch}}} at {{{worktree}}} for code standards compliance. Repo remote: {{{repoRemote}}}.'
-    failed:
-      role: $END
-      prompt: 'Developer failed: {{{reason}}}. Ending workflow.'
-  reviewer:
-    rejected:
-      role: developer
-      prompt: 'Reviewer rejected: {{{comments}}}. Fix the issues in repo {{{worktree}}}. Repo remote: {{{repoRemote}}}.'
-    approved:
-      role: tester
-      prompt: 'Review passed. Run tests on branch {{{branch}}} at {{{worktree}}}. Repo remote: {{{repoRemote}}}.'
-  tester:
-    fix_code:
-      role: developer
-      prompt: 'Tests found code issues: {{{report}}}. Fix and re-submit. Worktree: {{{worktree}}}. Repo remote: {{{repoRemote}}}.'
-    fix_spec:
-      role: planner
-      prompt: 'Tests found spec issues: {{{report}}}. Revise the test spec. Repo remote: {{{repoRemote}}}.'
-    passed:
-      role: committer
-      prompt: 'All tests passed. Commit and push branch {{{branch}}} from {{{worktree}}}. Repo remote (owner/repo): {{{repoRemote}}}.'
-  committer:
-    hook_failed:
-      role: developer
-      prompt: 'Push hook failed: {{{error}}}. Fix and re-submit. Worktree: {{{worktree}}}. Repo remote: {{{repoRemote}}}.'
-    committed:
-      role: $END
-      prompt: 'PR created: {{{prUrl}}}. Workflow complete.'