agentleFS
allemaarallemaar/open-skills/skills/insight-critique/SKILL.md

insight-critique

Skill15 starsChanged 16 days agoApache-2.0View on GitHub

Advisory review workflow. Reviews the last plan, code, architecture, or UI output and returns a structured report: Summary, Strengths, Risks, Fixes. Use when the user runs /critique, asks to "review this", "critique this", or "what's wrong with this", or wants focused feedback on a specific output. Also trigger as a quality gate before /execute. Advisory only; applicable controlling authority and safety rules still govern. For multi-POV adversarial stress-testing with personas, use /insight-adversarial instead.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run all four passes. Passes 1–3 are independent.

    line 38
  2. Distinctive:

    Never mark Approved when any risk is rated Medium or above.

    line 56
  3. Distinctive:

    Do not pad Strengths to soften the critique.

    line 57

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

/critique

Advisory review of the last plan, code, architecture, or UI output. Runs four analysis passes and emits a structured report. Advisory only; applicable controlling authority and safety rules still govern.

Structured execution spec: protocol.yon. Read it for the canonical rules and step sequence; this file is explanation. The two must stay in sync — if you edit one, update the other and refresh the @STAMP date.

Caller Options. Before executing, run the Caller Options protocol (caller-options/SKILL.md): triage this invocation for material optionality across the venues and modes declared in front-matter; if one path clearly dominates, proceed silently; otherwise surface the options to the caller. A resolved-invocation marker means COP already ran — execute the fixed combination directly, do not re-enter COP.

Phase 1 — Classify

Identify what is being critiqued. Pick exactly one: Code, Plan, Architecture, UI. If the subject is ambiguous, ask for clarification before proceeding.

Phase 2 — Analysis

Run all four passes. Passes 1–3 are independent. Pass 4 depends on Risks.

  1. Summary — what is this? State purpose, scope, and approach concisely.
  2. Strengths — what is well-designed, sound, or admirable? Be specific. Do not pad.
  3. Risks — what is fragile, dangerous, underspecified, or likely to fail? Rank by severity: High / Medium / Low.
  4. Fixes — for each risk, propose a concrete actionable change. Distinguish must-fix from nice-to-fix.

Phase 3 — Render Report

Output a structured report with four labeled sections: Summary, Strengths, Risks, Fixes. Then set a verdict:

  • ✅ Approved — all risks are Low or cosmetic
  • ⚠️ Conditional — risks present but fixable; propose targeted fixes; offer /plan-execute only against a relevant approved executable plan
  • 🚫 Not Approved — High risks present; /plan-create or a redesign is required

Constraints

  • Advisory only. Critique does not override applicable controlling authority or safety rules.
  • Never mark Approved when any risk is rated Medium or above.
  • Do not pad Strengths to soften the critique. If there are few strengths, say so.

Downstream

Suggest the natural next command — these mirror the next-skills: front-matter:

  • /insight-adversarial — stress-test the critiqued work from multiple POVs
  • /plan-create — if the verdict is 🚫 Not Approved and a redesign is needed
  • /plan-execute — only for a relevant approved executable plan after an Approved or Conditional verdict. Offer /plan-execute only when a relevant approved executable plan exists. Otherwise offer a scoped authorized repair or the appropriate planning owner; a critique verdict does not create execution approval.

Human output. This skill's handler-facing output obeys the human-output contract (human-output/SKILL.md).

Next skills. On completion, run the Next Skills protocol (next-skills/SKILL.md): surface the next-skills recommendations from front-matter for the caller to pick. Offer only — never auto-invoke.

Self-improvement. On completion, run the Self-Improvement Protocol (self-improve/SKILL.md): apply its signal taxonomy and threshold to this run; surface any surviving proposal for handler approval. Conservative — silent otherwise. Never auto-apply.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.