agentleFS
fireshortfireshort/skills/writing-plans/SKILL.md

writing-plans

Skill1 starsChanged 3 months agoMITView on GitHub

Use when requirements or a technical spec are settled and a multi-step change needs an actionable file-level implementation sequence, dependency ordering, or explicit verification. Do not use for simple localized edits that can be implemented and validated safely without a separate plan.

187 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use chinese-code-review, chinese-commit-conventions and 10 more.

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.
  • Leodorareluctant259Skill

    chinese-code-review

    中文代码审查规范——在保持专业严谨的同时,用符合国内团队文化的方式给出有效反馈

    Also in 8 projects that use this

  • Leodorareluctant259Skill

    chinese-commit-conventions

    中文 Git 提交规范 — 适配国内团队的 commit message 规范和 changelog 自动化

    Also in 8 projects that use this

  • ryanportfolioSkill

    addskill

    Create, import, update, or install repo or personal skills: runtime ownership, resources, discovery checks.

    Also in 6 projects that use this

  • ryanportfolioSkill

    apply-firmware

    Bring current folder (plain folder, existing repo, old or partial firmware copy) up to full Harness Firmware in place. Mirror external repo → adopt-repo.

    Also in 6 projects that use this

  • +2 more from Leodorareluctant259
  • +6 more from ryanportfolio

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 169 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:

    Read project rules, the governing brief or spec, and the actual source, configuration, tests, scripts, and downstream consumers affected by …

    line 18
  2. Distinctive:

    If there is no real multi-step sequence, dependency, validation boundary, ownership boundary, or rollback boundary, stop without producing …

    line 19
  3. Distinctive:

    Use exact commands when they are known from the project.

    line 22
  4. Distinctive:

    Add rollout, compatibility, data migration, rollback, or documentation work only when the change creates that obligation.

    line 23
  5. Distinctive:

    Keep a single cohesive edit together when splitting it would add coordination without reducing risk.

    line 27
  6. Distinctive:

    Prefer targeted tests and smoke checks first, then broader validation when the impact justifies it.

    line 30
  7. Distinctive:

    Before handing off the plan, check that:

    line 73
  8. Distinctive:

    Every task traces to a settled requirement or design decision.

    line 75

“Common” is measured: kept by at least 6 in 10 of the 169 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

Writing Plans

Purpose

Turn settled requirements or an approved spec into a proportional implementation plan. A plan tells an implementer what to change, in what order, and how to prove it works. It does not reopen product or architecture decisions and does not expand every obvious edit into minute-by-minute steps.

Default Location

Save plans to docs/plans/YYYY-MM-DD-<topic>.md unless project rules, an existing convention, or the user specifies another location. Preserve existing historical paths rather than renaming them for this skill.

Workflow

  1. Read project rules, the governing brief or spec, and the actual source, configuration, tests, scripts, and downstream consumers affected by the change.
  2. If there is no real multi-step sequence, dependency, validation boundary, ownership boundary, or rollback boundary, stop without producing a placeholder plan and recommend direct implementation.
  3. Confirm that no blocking requirement or design decision remains. If implementation would decide product scope, a public contract, security, migration, or architecture, return to the appropriate brief or spec instead of hiding the decision in the plan.
  4. Define the smallest safe sequence of tasks. Name target files or discoverable locations, describe the intended change, capture dependencies, and attach a concrete verification step to each meaningful task.
  5. Use exact commands when they are known from the project. Do not invent commands, APIs, file paths, or test facilities; mark discovery or manual validation where needed.
  6. Add rollout, compatibility, data migration, rollback, or documentation work only when the change creates that obligation.
  7. Self-review the plan and follow the project's review rules. Do not automatically start implementation, create a worktree, delegate to subagents, require test-first mechanics, or present branch-finishing choices.

Planning Granularity

  • Keep a single cohesive edit together when splitting it would add coordination without reducing risk.
  • Split tasks at real dependency, validation, ownership, or rollback boundaries.
  • Include code snippets only when an exact contract or non-obvious algorithm must be preserved; otherwise describe intent and point to the relevant symbols.
  • Prefer targeted tests and smoke checks first, then broader validation when the impact justifies it.

Plan Template

# <Change> - Implementation Plan

- Date: YYYY-MM-DD
- Status: Plan pending review
- Inputs: <brief, spec, issue, or request>

## Goal
<The outcome this plan will deliver.>

## Scope
- In scope: <files, components, or behaviors>
- Out of scope: <explicit exclusions>

## Tasks

### 1. <Meaningful implementation unit>
- Files: `<path>` and relevant symbols
- Change: <what to add, remove, or preserve>
- Depends on: <earlier task or external condition, if any>
- Verify: `<targeted command>` or <specific manual/runtime check>

### 2. <Next meaningful implementation unit>
- Files: `<path>`
- Change: <intended behavior>
- Verify: <evidence required>

## Final Validation
- <Integrated regression, build, runtime, browser, or client check proportionate to risk>

## Rollout and Rollback
<Include only when deployment, migration, compatibility, or irreversible state warrants it.>

## Open Blockers
<Only issues that prevent safe implementation.>

Quality Bar

Before handing off the plan, check that:

  • Every task traces to a settled requirement or design decision.
  • File paths, symbols, commands, and dependencies were checked against the real project.
  • Each meaningful task has evidence that can show whether it succeeded.
  • The sequence exposes risky changes early and avoids unrelated refactoring.
  • The plan is detailed enough to execute without becoming a transcript of trivial edits.

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.