cskwork/supergoal-skill / skills/supergoal
supergoal
UnverifiedRoute objective-driven development, verification, planning, "teach/explain", codebase learning, and skill evaluation through the Supergoal workflow. Use when Supergoal is requested or an autonomous delivery workflow is needed.
reference/review-only.md
View on GitHub# REVIEW-ONLY mode - findings, not fixes
Use when the user wants a code review or audit and explicitly wants no changes: "review this
code/diff/PR", "audit this module", "find issues", "코드 리뷰만", "고치지는 말고 봐줘". The deliverable
is an evidence-backed findings report. It writes NO source or test edits - read-only except the run
folder. If the user wants the findings fixed, that is a new objective: route to DEBUG/LEGACY with the
report as the Frame input.
## Frame (one line)
Name the review target (working diff, branch vs base, PR number, or files), the base it is compared
against, and the depth the user asked for (quick pass vs thorough audit).
## Dispatch - two independent reviewers in parallel
- `agents/code-reviewer.md` in findings-only stance: correctness, test adequacy, readability, error
handling, dead code. In this mode it does NOT write failing test files; each untested required
behavior becomes a finding that names the missing test it would write (file, behavior, edge).
- `agents/security-reviewer.md`: secrets, injection, SSRF/XSS, auth, unsafe crypto, input validation.
Both read request/docs and repo/data rules (`reference/domain-context.md`, `domain-rules.md`) so
findings are judged against what the project requires, not generic style taste.
## Verify findings before reporting
Every CRITICAL/HIGH finding is re-checked against the cited code (re-read the lines; run the cited
test or command when one exists) before it enters the report. Drop or downgrade anything you cannot
back with evidence - a plausible-but-unverified finding is the failure mode. Findings never override
a passing real test; they explain why the test is insufficient instead.
## Report (the one deliverable)
Write `report.md` in the run vault (`docs/changelog/<YYYY-MM>/<DD-review-topic>/`), severity-ordered:
- `Target:` what was reviewed and against which base.
- Findings grouped CRITICAL / HIGH / MEDIUM / LOW: each carries file:line, what is wrong, why it
matters here, and a concrete fix suggestion (suggestion only - not applied).
- `Untested behaviors:` the missing-test findings from the critic stance.
- `Not covered:` what the review did not look at, so silence is not read as approval.
## Exit
Report delivered; repo untouched (`git status` clean except the vault). Offer the route: "to fix
these, run DEBUG/LEGACY with this report as input" - do not start fixing in this mode.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |