recompose: Skill for Claude Code

.claude/skills/feature-cycle/SKILL.md

feature-cycle is a skill for Claude Code from recomposesh/recompose. It costs 62 tokens per session (2,083 once invoked), scanned B, original, MIT.

A process for taking a feature request from initial idea through discovery, design, implementation, review, and a pull request. A pull request is a proposed code change submitted for review before merging.

In plain words
What is it for?
Running an end-to-end feature workflow, creating its proposal and specifications, implementing tasks, reviewing the result, and preparing a pull request.
Why use it?
It gives feature work a defined sequence, documents the work on disk, and places human approval points between stages. It coordinates supporting processes for planning and task-by-task implementation.

Skill for Claude Code

Written for Claude Code: PreToolUse hook event. Also seen: reads .claude/ paths; mentions subagents.

This is recomposesh/recompose's own configuration. It tells Claude Code how to work on recompose itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything recompose configures →

Reuse

Borrowing it

Nothing to install: this file belongs to recomposesh/recompose. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/recomposesh/recompose/main/.claude/skills/feature-cycle/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/recomposesh/recompose

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for feature-cycle

README.md
[![agentmods](https://agentmods.dev/badge/skills/recomposesh/recompose/feature-cycle/github.svg)](https://agentmods.dev/skills/recomposesh/recompose/feature-cycle)
Your own site
<a href="https://agentmods.dev/skills/recomposesh/recompose/feature-cycle"><img src="https://agentmods.dev/badge/skills/recomposesh/recompose/feature-cycle/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for feature-cycle

Your own site · 80×15
<a href="https://agentmods.dev/skills/recomposesh/recompose/feature-cycle"><img src="https://agentmods.dev/badge/skills/recomposesh/recompose/feature-cycle.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 62 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,083 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00062 $0.02083
Opus 5.5 $0.00025 $0.00833
Sonnet 5.5 $0.00012 $0.00417
Haiku 4.5 $0.00006 $0.00208

Measured 1mo ago against content hash 34fb827ce2b0, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade B, and why

feature-cycle scanned grade B with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

The phases name enforcement machinery. The finding-by-commit verifiers stay deferred until a feature has run through the pipeline, because building them before a round has produced findings would fix their shape against
.claude/skills/feature-cycle/SKILL.md · 102 lines

How it starts

The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Feature cycle

The feature cycle turns a feature idea into a merged pull request through a graph of subagent nodes, deterministic code nodes, and human gates. This skill is the process definition. It orchestrates two supporting layers and owns everything between them.

  • OpenSpec owns the artifact layer. openspec new change <slug> scaffolds openspec/changes/<slug>/ with the recompose schema slots: proposal, specs, design, tasks, discovery, manifest, and gherkin. Merge archives the deltas into openspec/specs/.
  • Superpowers is the execution library. Implementation delegates task-by-task execution to superpowers:subagent-driven-development and its resume ledger; planning borrows the brainstorming question discipline.
  • This skill owns the graph, the tier rubric, the gate rules, the convergence rules, and the artifact contracts.

Human gates split the pipeline into separate runs that share no live state. Each run reads its inputs from the change directory and manifest.md, then writes its outputs back. The artifacts on disk are the edges between runs.

Entry contract

Trigger: /feature-cycle <description>. On entry, before any phase work:

  1. Classify. A Haiku classifier fills isUI and the affected subsystems, then recommends trivial, standard, or full with its rubric reasons.
  2. Confirm the tier. The maintainer confirms or overrides in one word. This is recommend-then-confirm: the classifier proposes, the maintainer decides.
  3. Scaffold the change and seed a real delta in one commit. See Change hygiene.
  4. Run the phase set for the confirmed tier. The reference files carry the detail: planning.md, implementation.md, verification.md.

A confirmed trivial tier exits after step 2 with no change directory scaffolded. Steps 3 and 4 run only for standard and full.

Tier rubric

The recommendation follows a written rubric. The maintainer confirms it.

Read the full file on GitHub · 102 lines

Files

What ships with it

3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 1mo ago First seen · 102 lines · 62 tokens per session scan B 34fb827ce2b0

Subscribe to this mod's changes

feature-cycle is a skill published in the GitHub repository recomposesh/recompose (30 stars, last pushed 1mo ago), licensed MIT. It adds 62 tokens to every session and 2,083 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.