agentleFS
tornikebolokadze1-cybertornikebolokadze1-cyber/agent-foundation-kit/.claude/skills/spec/SKILL.md

spec

Skill6 starsChanged 27 days agoMITView on GitHub

Write or revise a feature specification (docs/specs/SPEC-NNNN-slug.md) from the accepted intent, the domain facts and the accepted ADRs, then have it reviewed by the spec-reviewer subagent. Use when the owner wants a feature defined, or before planning any feature.

176 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use ios-clean, ios-fix 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.
  • 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

  • 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

  • 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

1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 152 edited versions we read keep these.

  • 1

    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:

    Success measures (§6, ADR-0013): Ask the person, in plain words, „რით გავიგებთ, რომ მუშაობს?“ — one question, 2–3 options with your …

    line 27

“Common” is measured: kept by at least 6 in 10 of the 152 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 — a specification the owner can approve and a machine can check

Precondition: intent.md is accepted. If it is not, stop and tell the owner that /approve intent comes first (a spec cannot be approved before that — approve.py refuses and the CHAIN gate fails).

  1. Context — goals G* and non-goals in intent.md, docs/domain/*.md, accepted ADRs (docs/adr/README.md), existing specs; bash .claude/hooks/py.sh tools/context_graph.py query intent.
  2. New file — next free number SPEC-NNNN; docs/specs/SPEC-NNNN-<english-kebab-slug>.md from template.md. Revising an approved spec: reopen it first (status only → draft).
  3. Fill every section. Requirements REQ-NNN each trace to a goal G*. Acceptance criteria are Given / When / Then, testable, and each names the gate, test or eval that will prove it — and, in the column „რას დაინახავ ან გააკეთებ", what the person will see or do to check it with their own eyes („ტელეფონზე გახსნი გვერდს და პირველ ეკრანზე ღილაკს დაინახავ"); a criterion nobody can see says so in words („თვალით არ ჩანს — ამას ტესტი ამტკიცებს"), never an empty cell or a dash. Include Georgian-language quality, grounding and the domain invariants (REVIEW.md, docs/domain/) where they apply. Anything unknown → "ღია კითხვები", never a guess. The first version's scope and example criteria for this kind of project are in docs/guides/<kind>.md §3. Keep a first version's spec small enough for a beginner to judge: about 5–8 requirements and at most ~10 acceptance criteria; ask a missing fact in the chat (one question per message) rather than piling up open questions — at most 3 stay in the file. Describe what the result does, not how it is built: the stack is decided at stage 3. Success measures (§6, ADR-0013). Ask the person, in plain words, „რით გავიგებთ, რომ მუშაობს?“ — one question, 2–3 options with your recommendation (answers right on a sample, speed, cost per request, no wrong price on the page…) — and write the answer as 2–4 measures M-NNN, each with a threshold, where it is measured (a test, an eval set in evals/, a gate) and when (every change; weekly at stage 6). For an AI product these thresholds are the level-2 eval gates' contract (evals-georgian, evals-grounding): the gates are switched on at stage 4 with the person's yes and their thresholds never go down; the eval set is written with the spec, not after the code.
  4. Index — add a row to docs/specs/README.md (wikilink to the file). When complete, set status: review.
  5. Review — run the spec-reviewer subagent on the file. Fix critical/high findings; record the rest under "ღია კითხვები". Then the builder check (builder-check.md) on the spec, its questions sorted into the same three piles as in /intent step 4: the person's pile — at most three, in one message — goes with the summary of step 7, and their answers go into the spec before they approve it.
  6. Check — bash .claude/hooks/py.sh tools/context_graph.py check <file>; fix every error. From review on it also fails on a line left from the template, an empty cell in the requirements or criteria table, or more than 3 open questions.
  7. Tell the owner — 3–5 Georgian sentences: what the spec covers, what is out of scope, what they must decide; this summary is what they judge — never ask them to read the file. The approval is /approve spec NNNN. Planning (/feature-plan) starts only after it.

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.