agentleFS
sainzssainzs/santiagosainz-skills/skills/core/implementation-plan/SKILL.md

implementation-plan

Skill1 starsChanged 3 months agoMITView on GitHub

Convert a plan, PRD, backlog item, or vague feature into ordered implementation work with acceptance criteria and verification. Use when the user asks to break down work, make an implementation plan, create tasks, plan execution, or use slice-plan.

25 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use code-quality and security-review.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • MicrosoftSkill

    code-quality

    Statically check changed C# against the same deterministic rules github-code-quality[bot] runs on pull requests, so a contributor can fix likely findings before opening the PR and keep the bot's inline comments to a minimum. Activate when a contributor asks to "check code quality", "run code quality on my changes", "will the code quality bot complain", "pre-check before I open the PR", "code quality check", or similar. Scoped to C#. Reads the diff and compares it against the enumerated csharp-code-quality rule set below — no CodeQL download or build required. This is a fast approximation of a dataflow engine, not a reproduction; report accordingly. Does NOT apply fixes unless explicitly asked.

    Also in 3 projects that use this

  • SentrySkill

    security-review

    Finds exploitable application security vulnerabilities in code changes. Use for Warden security scans, appsec review, OWASP-style checks, authentication or authorization bugs, injection, XSS, SSRF, path traversal, secrets, unsafe crypto, webhook verification, open redirects, or sensitive data exposure.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 23 edited versions we read keep these.

  • 8

    Distinctive

    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:

    Do one job: convert intent into ordered, independently provable work.

    line 12
  2. Distinctive:

    Avoid ceremony: no full process framework when a two-step plan is enough.

    line 15
  3. Distinctive:

    Keep tool and tracker assumptions explicit.

    line 16
  4. Distinctive:

    Use for plan breakdowns, implementation sequencing, issue creation, PRD-to-work conversion, or when a task is too large to complete safely …

    line 20
  5. Distinctive:

    Prefer vertical tasks through the real interface over layer-by-layer tasks.

    line 26
  6. Distinctive:

    Keep dependencies explicit and minimal.

    line 28
  7. Distinctive:

    Read the source plan and the current implementation surface.

    line 32
  8. Distinctive:

    If the user asked to create issues or files, publish/write them after confirming any tracker or path requirements.

    line 36

“Common” is measured: kept by at least 6 in 10 of the 23 edited versions of this file we could read. “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

implementation-plan

Break work into tasks that can be built, reviewed, and verified independently.

Unix/Pi Contract

  • Do one job: convert intent into ordered, independently provable work.
  • Consume a plan, PRD, issue, or conversation; emit tasks with acceptance criteria and verification.
  • Make every task pipeable into PLAN.md, a tracker issue, or an implementation prompt.
  • Avoid ceremony: no full process framework when a two-step plan is enough.
  • Keep tool and tracker assumptions explicit.

Use

Use for plan breakdowns, implementation sequencing, issue creation, PRD-to-work conversion, or when a task is too large to complete safely in one edit.

Slice Rules

  • Each task should produce observable behavior or a meaningful invariant.
  • Each task should include tests or a concrete verification command.
  • Prefer vertical tasks through the real interface over layer-by-layer tasks.
  • Mark human-decision tasks separately from agent-executable tasks.
  • Keep dependencies explicit and minimal.

Workflow

  1. Read the source plan and the current implementation surface.
  2. Identify the narrowest first task that reduces uncertainty.
  3. Split remaining work by user-visible behavior, contract, or migration step.
  4. Attach acceptance criteria, files likely touched, and verification commands.
  5. If the user asked to create issues or files, publish/write them after confirming any tracker or path requirements.

Output

For each task:

  • Title.
  • Type: executable, decision, review, or cleanup.
  • Acceptance criteria.
  • Likely files.
  • Verification.
  • Dependencies.

End with the recommended first task.

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.