agentleFS
dtiger1889-opsdtiger1889-ops/claude-harness-toolbox/skills/spec/SKILL.md

spec

Skill31 starsChanged 31 days agoMITView on GitHub

Writes or reviews a change or decision spec with explicit tradeoffs and closeout criteria. Use for /spec, "spec this", "write a spec", review of a spec's open questions, or closing, archiving, and promoting a finished spec. Does not authorize implementation by itself.

168 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use backprop, ios-clean 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.
  • JuliusBrusseeSkill

    backprop

    Bug → spec protocol. When a bug is found or a test fails, trace the cause, decide whether a new §V invariant would catch recurrence, append to §B. This is the one non-obvious thing SDD does that plan-then-execute doesn't. Triggers on test failure, bug report, post-mortem, or explicit user ask.

    Also in 4 projects that use this

  • Saqlain-811Skill

    ios-clean

    Remove the DebugBridge SPM package and all #if DEBUG wiring from an iOS app. (gstack)

    Also in 5 projects that use this

  • Saqlain-811Skill

    ios-fix

    Autonomous iOS bug fixer. (gstack)

    Also in 5 projects that use this

  • bastosSkill

    ios-design-review

    Review iOS app UI/UX design quality using Apple's design principles and the Liquid Glass-era design language from WWDC25/26. Use when asked for a design review, UX critique, visual design assessment, or evaluation of an app's interface quality, design intent, alignment with the new Apple design system, or usability. Triggers: "design review", "UX review", "visual design critique", "review the UI", "does this design work", "evaluate this interface", "design feedback", "Liquid Glass review".

    Also in 5 projects that use this

  • ShreyRaviSkill

    plan-tune

    Self-tuning question sensitivity + developer psychographic for gstack (v1: observational). Review which AskUserQuestion prompts fire across gstack skills, set per-question preferences (never-ask / always-ask / ask-only-for-one-way), inspect the dual-track profile (what you declared vs what your behavior suggests), and enable/disable question tuning. Conversational interface — no CLI syntax required.

    Also in 7 projects that use this

  • timurgaleevSkill

    document-generate

    Generate missing docs for a feature, module or project as tutorials, how-tos, reference and explanation.

    Also in 6 projects that use this

  • +2 more from Saqlain-811

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 6 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 145 edited versions we read keep these.

  • 6

    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 the project's current state, instructions, existing spec, and relevant artifacts.

    line 8
  2. Distinctive:

    Do not invent alternatives for a question with one answer.

    line 27
  3. Distinctive:

    Verify completion against actual artifacts and acceptance checks.

    line 61
  4. Distinctive:

    If a live change still needs release, use its authorized project release path.

    line 62
  5. Distinctive:

    Update the state/tracker pointer in the same turn.

    line 64
  6. Distinctive:

    If the runtime cannot move files, preserve the source and record the limitation.

    line 66

“Common” is measured: kept by at least 6 in 10 of the 145 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

Spec

Read the project's current state, instructions, existing spec, and relevant artifacts. Follow its naming and storage conventions. Otherwise use the existing outbound/ or specs/ folder, or create specs/. Write one topic per file and link it from the project's state or tracker.

Write or review

Choose depth by the work:

  • Change spec: concise problem, intended behavior, scope, implementation plan, and acceptance checks.
  • Decision spec: enough evidence to assess the recommendation without repeating the investigation: methods, measurements, rejected approaches, uncertainty, costs, and sources.

For later execution, put the plan near the top. Name exact files, relevant anchors or patterns, configuration defaults, disable/revert path, tests, and acceptance checks. Prefer stable symbols or headings to unexplained line numbers.

Attribute choices accurately: user decisions, measurements, and model recommendations are different. Preserve exact user wording only when it defines a requirement or decision; do not quote every exchange. State risks, affected surfaces, and non-goals. Use an actual date and avoid invented shorthand.

For each unresolved choice the user must make, include:

  • The question and necessary context.
  • Two or three real options, each with gains, losses, implementation cost, and maintenance cost.
  • A recommendation, the one piece of evidence it rests on (file and line, command output, URL, or the user's own words with their date) or "no evidence beyond judgment", and the condition that would change it. A recommendation with a confident reason and no source reads as a finding when it is a guess.

Do not invent alternatives for a question with one answer. Support claims that an option fixes a failure with the relevant issue, test, or failure record; otherwise mark that claim unverified.

File shape

Adapt sections to the task; omit empty sections except Closeout.

# <Topic>
Status: proposed
Created: <date>
Owner / attribution: <who made decisions or recommendations>

## Execution plan
<when implementation is intended>

## Problem and goal
## Scope and non-goals
## Evidence and rejected approaches
<for investigated decisions>
## Design
## Choices requiring the user's decision
<options, gains, losses, costs, recommendation, its evidence, reversal condition>
## Acceptance checks
<observable success and relevant negative cases>
## Closeout
- [ ] Verify the work or decision against the acceptance checks.
- [ ] Release any live artifact through the project's authorized release path.
- [ ] Move this spec to decisions/ or archive/, update Status, and retire the open pointer.

Report the path and status briefly. A spec request alone authorizes the document, not the build. Honor explicit existing implementation authorization; do not reset it merely because a spec is being written.

Close or promote

  1. Verify completion against actual artifacts and acceptance checks. Report unresolved work as open.
  2. If a live change still needs release, use its authorized project release path. Do not infer permission for a new external action from "close the spec".
  3. Move the completed spec to decisions/ or archive/ using a real file move. Record whether it was applied or is only a decision record.
  4. Update the state/tracker pointer in the same turn. Confirm the outcome and new path.

If the runtime cannot move files, preserve the source and record the limitation. A copied destination with a surviving original is not a completed archive.

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.