Stefano Fiorini
f404792927
Ports the claude-code canonical to Cursor Agent CLI conventions: - Workspace skill discovery (.cursor/skills/ repo-local preferred, ~/.cursor/skills/ global fallback) replaces Skill-tool invocations. - Helper path resolution prefers .cursor/skills/reviewer-runtime/ over ~/.cursor/skills/reviewer-runtime/. - jq added as a hard prerequisite; cursor-agent --version check added. - Plan-mode guard removed (Cursor has no plan-mode concept). - Reviewer invocations mandated to --mode=ask --trust --output-format json; explicit ban on --mode=agent and --force. - Added Required Skill Invocation Rules, Variant Hardening Notes — Cursor, Common Mistakes, and Red Flags sections. - Template runtime field = cursor; guardrail line updated. Core-section schema identical to claude-code canonical. Reviewer: codex / gpt-5.4. Approved round 1 (0 P0/P1/P2/P3). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
4.4 KiB
4.4 KiB
Task Plan: [Short Title]
Variant guardrail (Cursor): Sub-skills (
brainstorming,test-driven-development,verification-before-completion,finishing-a-development-branch,using-git-worktrees) MUST be invoked through workspace discovery from.cursor/skills/superpowers/skills/<skill>/SKILL.mdor~/.cursor/skills/superpowers/skills/<skill>/SKILL.md. Reviewer invocations MUST use--mode=ask --trust --output-format json.jqis a hard prerequisite.
Metadata
| Field | Value |
|---|---|
| Created | YYYY-MM-DD |
| Slug | YYYY-MM-DD- |
| Runtime | cursor |
| Reviewer CLI | codex | claude | cursor | opencode |
| Reviewer Model | |
| MAX_ROUNDS | 10 |
| Branch Strategy | current-branch | worktree |
| Branch Name | <current branch name, or new branch name when worktree is used> |
| Worktree Path | <absolute path to worktree dir; blank when Branch Strategy = current-branch> |
| Status | draft |
Status Enum (authoritative)
| Value | Meaning |
|---|---|
draft |
Newly created; plan review not yet started |
plan-approved |
Plan review loop returned APPROVED |
implementation-in-progress |
Phase 6 executing |
implementation-approved |
Phase 8 review loop returned APPROVED; awaiting commit |
pushed |
Committed + pushed to remote |
local-only |
Committed locally; user declined push |
aborted-plan-review |
MAX_ROUNDS reached in Phase 5; user aborted |
aborted-impl-review |
MAX_ROUNDS reached in Phase 8; user aborted |
aborted-verification |
Phase 7 retries exhausted; user aborted |
failed |
Hard tooling failure |
Prompt
Interpretation
Assumptions
Files
| Action | Path | Why |
|---|---|---|
Approach
TDD Approach
Acceptance Criteria
- <criterion 1>
- <criterion 2>
Verification
Rollback
Runtime State
plan_review_round: 0
implementation_review_round: 0
CODEX_PLAN_SESSION_ID:
CODEX_IMPL_SESSION_ID:
CURSOR_PLAN_SESSION_ID:
CURSOR_IMPL_SESSION_ID:
OPENCODE_PLAN_SESSION_ID:
OPENCODE_IMPL_SESSION_ID:
last_phase_entered:
last_round_ts:
last_scan_outcome_plan:
last_scan_outcome_impl:
verification_attempts: 0
tests_added_count: 0
tdd_used: false
Review History
| Timestamp (ISO-8601) | Loop | Round | Verdict | Summary |
|---|---|---|---|---|
Final Status
Guardrails (do NOT remove)
- This file is the single persistent artifact for
do-task. Do not split it or delete it on success. Statusmust always match one of the 10 enum values.Runtime Stateis updated by the skill, not by the user.- Review History is append-only.
last_scan_outcome_planandlast_scan_outcome_implrecord the most recent secret-scan result for each loop. They are informational; the scan itself runs per-payload with no caching.