agentleFS
IcodeNetIcodeNet/agent-skills/skills/explain-pr-step-by-step/SKILL.md

explain-pr-step-by-step

Skill0 starsChanged 4 months agoMITView on GitHub

Explain a pull request incrementally with concept teaching and confirmation gates between steps. Use when the user asks for a step-by-step PR walkthrough, what changed and why, or asks to pause after each step for deeper understanding.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 8 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    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:

    Use this skill to teach a PR in guided, checkpointed steps.

    line 8
  2. Distinctive:

    Do not infer chronology without checking commit/diff evidence.

    line 23
  3. Distinctive:

    Keep each step focused on one core idea.

    line 33
  4. Distinctive:

    If user asks "where?", "show example", "what does X mean?", pause progression.

    line 44
  5. Distinctive:

    Prefer "we" and "our" phrasing.

    line 49
  6. Distinctive:

    Avoid over-abstract explanations detached from repo code.

    line 50
  7. Distinctive:

    Always link concepts to concrete implementation, deriving examples from the PR's own diff.

    line 54
  8. Distinctive:

    Keep examples small and directly relevant to current files.

    line 59

≈ “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

Explain PR Step by Step

Use this skill to teach a PR in guided, checkpointed steps.

Goal

Produce a clear sequence of changes where each step includes:

  • what changed,
  • why it changed,
  • the concept behind it,
  • a pause question before moving on.

Required workflow

  1. Collect PR evidence first
    • Fetch PR metadata (title, summary/body, commits, files).
    • Fetch the full diff and per-file patches if needed.
    • Do not infer chronology without checking commit/diff evidence.
  2. Build a teaching sequence
    • Group changes into understandable phases, typically:
      1. feature flag/config enablement,
      2. contracts/types,
      3. state domain/data flow,
      4. UI renderer,
      5. runtime registration/wiring,
      6. tests and safety updates.
    • Keep each step focused on one core idea.
  3. Deliver one step at a time
    • For each step, include:
      • changed files/components,
      • behavioral outcome,
      • concept explanation (for example lifecycle, caching, hydration).
    • End each step with an explicit checkpoint question:
      • "Need more detail before we continue to Step N+1?"
  4. Honor drill-down questions immediately
    • If user asks "where?", "show example", "what does X mean?", pause progression.
    • Provide concrete code references and minimal examples.
    • Resume next numbered step only after user confirms.
  5. Use collaborative language
    • Prefer "we" and "our" phrasing.
    • Avoid over-abstract explanations detached from repo code.

Concept teaching rules

  • Always link concepts to concrete implementation, deriving examples from the PR's own diff. For example:
    • Feature flag -> rollout safety and gating behavior.
    • Schema/type change -> contract compatibility between producer and consumer.
    • New state field -> data flow and where the source of truth lives.
    • Lifecycle/status values -> what each state renders or triggers.
  • Keep examples small and directly relevant to current files.

Response template per step

Use this structure:

  1. Step N: <short title>
  2. What we changed (files and behavior)
  3. Concept (plain-language explanation)
  4. Why this matters
  5. Checkpoint question ("Need more understanding before Step N+1?")

Quality bar

  • Steps must be accurate to the diff.
  • Concepts must be technically correct and simple.
  • User must control pacing; never dump all steps unless asked.

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.