agentleFS
arxdsilvaarxdsilva/ai-loops/skills/feature/SKILL.md

feature

Skill0 starsChanged 7 days agoMITView on GitHub

Use when implementing a scoped backend feature or behaviour change in any language pack (new endpoint, use case, rule, consumer or job), or adding a package, module or command. Covers ownership, TDD, layering, consistency and the completion gate.

75 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use bug, bugfix 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.
  • paulocsbSkill

    bug

    Investigation playbook for fixing a bug. Walks through 5 phases: reproduce → isolate → root cause → propose fix → validate. Forces concrete reasoning with code reads, log inspection, and git blame instead of pattern-matching guesses. Use for any non-trivial defect.

    Also in 8 projects that use this

  • team-attentionSkill

    bugfix

    Root cause based one-shot bug fix. debugger diagnosis → requirements.md generation → /execute. /bugfix "error description" Full investigation pipeline: debugger + gap-analyzer + standard verify. QA suggestion after successful fix.

    Also in 7 projects that use this

  • codebycoreySkill

    adopt

    Adopt an existing plan from conversation context into delta-spec format. Use when importing a plan, adopting planning that already happened, capturing plan mode output, or bringing in prior design work.

    Also in 4 projects that use this

  • Google CloudSkill

    fix

    Plan a remediation strategy after a test failure.

    Also in 10 projects that use this

  • marylinSkill

    recover

    Recover work from a crashed or interrupted session. Use when Claude crashed mid-task, after an unexpected exit, when you see 'stale sessions detected', or when you lost context and need to figure out what was happening.

    Also in 3 projects that use this

  • aegiswizardSkill

    context

    Also in 4 projects that use this

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 2 are distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 69 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Write or update the failing behavioural test first.

    line 21
  2. Distinctive:

    Add wiring only where the feature crosses a port.

    line 23
  3. Very specific:

    Check consistency (skills/consistency) and compatibility: API, persisted schema, events, config, migrations.

    line 24
  4. Very specific:

    Run the pack's checks on your scope: loops gate --only <check>,… --pkgs <owned> while iterating, then loops gate --pkgs <owned> (the race …

    line 25
  5. Very specific:

    Update the package AGENTS.md if a contract changed (skills/documentation).

    line 26
  6. Very specific:

    Create one only when a task needs it.

    line 31

“Common” is measured: kept by at least 6 in 10 of the 69 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

Feature implementation

Specs live at the absolute path in your prompt (loops path specs). Cite by specs/<FILE>.md §N.

Load

  • Root AGENTS.md, the nearest package AGENTS.md, the public contracts and existing tests of the behaviour.
  • Layering and ports: specs/ARCHITECTURE.md §3, §6. Definition of done: specs/WORKFLOW.md §8.
  • Context rules: specs/CONTEXT_DISCIPLINE.md §8-10.
  • Your language (Language: in your prompt): loops:lang-<id> for the layout, commands and idioms.

Steps

  1. Restate the behaviour as observable acceptance criteria.
  2. Identify the owning package; search for existing canonical types and primitives before adding any (Go: specs/languages/go/RULES.md §48).
  3. Write or update the failing behavioural test first.
  4. Implement the smallest complete change in the right layer, removing before adding and keeping each decision in one place (specs/ENGINEERING.md §1, §3); model illegal states out and parse input once at the boundary (§8): business rules in the domain layer, orchestration in the application layer, technology in the adapters (the pack's layout tree; Go: internal/domain, internal/application, internal/adapters). No business rules in HTTP handlers, queue consumers, SQL repositories or entry points.
  5. Add wiring only where the feature crosses a port; construct concrete types at the composition root.
  6. Check consistency (skills/consistency) and compatibility: API, persisted schema, events, config, migrations.
  7. Run the pack's checks on your scope: loops gate --only <check>,… --pkgs <owned> while iterating, then loops gate --pkgs <owned> (the race policy is the pack's, WORKFLOW §6); benchmark if a hot path changed (skills/performance).
  8. Update the package AGENTS.md if a contract changed (skills/documentation).
  9. Finish with the claims report (templates/CLAIMS_REPORT.md).

New package or command

Create one only when a task needs it. Give it a AGENTS.md from templates/PACKAGE_AGENTS.md once it is real. Entry points are wiring only (specs/SOLID.md §8a; Go: cmd/<tool>/main.go); only the architect edits entry points and the composition root (Go: cmd/**, internal/compose/**; ROLES.md).

Answer before coding

Owner / State read / State written / Transaction boundary / Update frequency / Ordering
Index or query used / Allocation behaviour / Persistence / Idempotency / Config surface / Observability

Avoid

New global registries; transport-driven domain mutation; refactors unrelated to the acceptance criteria; duplicate IDs or types; global scans where an index exists; a thread, task or goroutine per item; wall-clock in domain; decisions from hash-map iteration order; a framework for one feature.

Completion gate

specs/WORKFLOW.md §8 and: failure paths explicit and tested; compatibility addressed; benchmarks exist or stay valid for hot changes; loops arch shows no errors.

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.