dynos-work: Status

Show the current state of the active dynos-work task.

Ruthlessness Standard

• Missing evidence is not success. Report absent artifacts as gaps.
• Do not infer PASS from silence, skipped files, or stale reports.
• The status report must make blockers and weak spots obvious, not polite.

What you do

1. Find the most recent active task in .dynos/ (manifest.json with stage not DONE/FAILED). If available in this repo, use "${CODEX_PLUGIN_ROOT:-${CLAUDE_PLUGIN_ROOT:-}}/bin/dynos" ctl active-task.
2. If no active task, report "No active dynos-work task found. Start one with /dynos-work:start"
3. Read: manifest.json, spec.md, execution-graph.json, latest audit-reports, repair-log.json, test-results.json, execution-log.md
4. Run "${CODEX_PLUGIN_ROOT:-${CLAUDE_PLUGIN_ROOT:-}}/bin/dynos" ctl task-churn .dynos/task-{id} and copy its counts and warnings into the Churn block. These are the loops operators could not see: re-approvals after post-approval edits, gate refusals, write-policy denials, planner attempts to run validators.
5. Print a human-readable status report

Output format

dynos-work Status Report
========================
Task: task-20260327-001
Title: [First 80 chars of task]
Stage: CHECKPOINT_AUDIT
Risk: medium
Snapshot: dynos/task-20260327-001-snapshot

Lifecycle Progress:
  ✓ DISCOVERY
  ✓ DISCOVERY
  ✓ DESIGN_REVIEW
  ✓ SPEC_NORMALIZATION
  ✓ SPEC_REVIEW (user-approved)
  ✓ PLANNING
  ✓ PLAN_REVIEW (user-approved)
  ✓ PLAN_AUDIT
  ✓ TDD_GATE
  ✓ PRE_EXECUTION_SNAPSHOT
  ✓ EXECUTION (3/3 segments complete)
  ✓ TEST_EXECUTION (all tests passed)
  → CHECKPOINT_AUDIT (in progress)
  ○ FINAL_AUDIT

Execution Progress:
  Segments: [N]/[total] complete
  Current batch: [seg-003-ui, seg-004-tests]
  Completed: [seg-001-db, seg-002-backend]

Acceptance Criteria: [N]/[total] covered

Churn (from ctl task-churn):
  Planner spawns: [N]   Approval receipts: [N]   Amendments: [N]
  Gate refusals: [N]    Write denials: [N]       Planner validator attempts: [N]
  Warnings: [each warning line, or "none"]

Latest Audit Results:
  spec-completion: PASS | FAIL | not yet run
  security: PASS | FAIL | not yet run
  ui: PASS | FAIL | SKIPPED | SKIPPED_REUSE | not yet run
  code-quality: PASS | FAIL | SKIPPED | SKIPPED_REUSE | not yet run
  db-schema: PASS | FAIL | SKIPPED | SKIPPED_REUSE | not yet run

Open Blocking Findings:
  [finding-id] [auditor]: [description] ([file:line])

Test Results: PASS | FAIL ([N] failing tests) | not yet run

Repair Cycle: [N] ([N] findings resolved, [N] remaining)

Next command: [one of the below based on current stage]

Recent Activity (last 10 lines of execution-log.md):
  [last 10 log entries, or "No execution log yet" if file doesn't exist]

Next command mapping

Always end the status report with a "Next:" line based on current stage: