agentleFS
q2522879285-sourceq2522879285-source/red-monkey-reasoning-kit/skills/reverse-engineer-example/SKILL.md

reverse-engineer-example

Skill2 starsChanged 2 months agoMITView on GitHub

Reverse-engineer a concrete strong example into reusable mechanisms, completion standards, an execution checklist, and a small transfer exercise. Use when the user provides an inspectable artifact or workflow and asks why it works or what can transfer. Skip unsupported success claims, copy requests, missing examples, film-specific breakdowns, and exact design replication.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 3 are distinctive, and 2 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.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use when an inspectable product page, workflow, document, dashboard, or other artifact is supplied and the user asks what makes it …

    line 12
  2. Very specific:

    Use $fact-claim-audit to verify performance or causal claims.

    line 27
  3. Very specific:

    Use $dual-layer-explanation only as a presentation layer for mechanisms already extracted.

    line 28
  4. Distinctive:

    Do not copy distinctive expression, private assets, or protected source content.

    line 29
  5. Distinctive:

    Every candidate rule points to a locatable observation, fact and inference are separate, the checklist is executable, and the exercise can …

    line 34

≈ “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

Reverse-Engineer Example

Own generic exemplar-to-mechanism distillation. Do not certify that the example caused its claimed outcome.

Route

Use when an inspectable product page, workflow, document, dashboard, or other artifact is supplied and the user asks what makes it effective, what rules transfer, or how to practice them.

Skip when there is no example, “excellent” is only promotional, the user wants a one-to-one copy, or the task is ordinary debugging or compliance review. Route film and video breakdowns to $film-breakdown-distiller; route exact design implementation to the applicable design-to-code skill.

Procedure

  1. Record the artifact, visible evidence, claimed results, transfer target, and unknown production variables.
  2. State in one sentence who it serves, what problem it solves, and the observable completion result.
  3. Reconstruct the structure or workflow and connect each pivotal choice as observation → candidate mechanism → expected effect.
  4. Extract completion standards; separate transferable mechanisms, domain constraints, case-specific expression, and unknown causality.
  5. Produce three to five conditional candidate rules with use conditions, failure conditions, and checks.
  6. Turn them into a minimal ordered checklist and one small exercise with a baseline, one main variable, and a result measure.

Handoffs and Boundaries

  • Use $fact-claim-audit to verify performance or causal claims.
  • Use $dual-layer-explanation only as a presentation layer for mechanisms already extracted.
  • Do not copy distinctive expression, private assets, or protected source content.
  • A single example yields hypotheses, not proven universal rules.

Completion Gate

Every candidate rule points to a locatable observation, fact and inference are separate, the checklist is executable, and the exercise can confirm or refute at least one transfer hypothesis.

Source Status

Distilled from the user-supplied “反向拆解” method card. The card offers a workflow but no verified result data, causal evidence, or transfer tests.

<!-- integration-watermark: 红猴子整合 -->

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.