feature

Orchestrate the full agentic-engineering chain on one brief — map → research → story → spec → rubric → backend lane → validator → frontend lane → final validator → PR — with three human checkpoints (story, spec+rubric, PR). Default path is in-conversation: each step runs as a forked Agent subagent so output is visible inline and per-step model selection is possible (Haiku for read-heavy steps, Sonnet/parent for writers, parent for code). Headless escape hatch: bash-invoke ${CLAUDE_SKILL_DIR}/bin/feature.sh for true AFK runs that survive terminal close. Triggers: /feature, "build this feature", "run the full chain", "orchestrate this feature end-to-end".

Quality

90%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

88%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A highly actionable, well-sequenced orchestrator body with concrete bash, exact state schema, explicit checkpoints, validation, and feedback loops. The main weakness is a monolithic single-file structure with minor over-explanation in a few sections.

Suggestions

Move the state.json schema and the per-step model-defaults table into reference files under references/ and link to them, so SKILL.md stays a lean overview.

Tighten the 'Honoring natural-language overrides' recipe and the redo-cap rationale — the reasoning behind cap-reset-on-resume can be trimmed without losing the rule.

Split the resume-state mapping table and failure-modes list into a references/resume-and-failure-modes.md to reduce inline bulk.

Dimension	Reasoning	Score
Conciseness	Dense with actionable specifics and justified 'Why' columns, but a few sections over-explain (redo-cap rationale, the multi-step model-override normalization recipe) and could be tightened.	4 / 5
Actionability	Provides copy-paste-ready bash (allowlist_for, route_findings), an exact state.json schema, and exact per-step subagent_type + model assignments — fully executable guidance covering the common cases.	5 / 5
Workflow Clarity	Steps 1–9 are explicitly sequenced with three labeled checkpoints, validation steps (allowlist validation, adversarial validators), feedback loops (re-dispatch with findings), and explicit loopback caps with persisted counters.	5 / 5
Progressive Disclosure	Good section structure with sub-skills referenced one level deep and clearly signaled, but the ~288-line SKILL.md is monolithic with no bundle reference files; the state schema and model-defaults table stay inline rather than being split out.	4 / 5
	Total	18 / 20 Passed

Description

92%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A specific, complete, third-person description that names the full chain of concrete actions, the checkpoint structure, and explicit natural-language triggers, with minimal conflict risk. It is somewhat verbose but the length is concrete rather than padded fluff.

Dimension	Reasoning	Score
Specificity	Enumerates the full concrete action chain (map → research → story → spec → rubric → backend lane → validator → frontend lane → final validator → PR) plus checkpoints, per-step model selection, and a headless escape hatch — comprehensive concrete actions.	5 / 5
Completeness	Explicitly answers both 'what' (orchestrate the full agentic-engineering chain) and 'when' via a dedicated 'Triggers:' clause with concrete phrases, in third person.	5 / 5
Trigger Term Quality	Explicit natural trigger phrases ('build this feature', 'run the full chain', 'orchestrate this feature end-to-end', '/feature') that users would actually say, though a few common synonyms/variations are missing.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche ('agentic-engineering chain' orchestration with '/feature') with distinct triggers and minimal overlap risk against other skills.	5 / 5
	Total	19 / 20 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: belchman/claude-skills
Path: plugins/agentic-engineering/skills/feature/SKILL.md
Commit: 77a9e6b

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.