implement-feature

Drive a researched and planned feature to a landed, tested change. Use after plan-feature has produced a docs/design/<project>/ workspace and the user says "implement it", "build the plan", "run the plan", or "let's ship this". Orchestrates refresh-plan, implement, review, a debug-local-deployment loop, and a test loop across the daytona / local-pi / claude x SDK / UI matrix, then documentation and a GitButler stacked branch. The orchestrator stays in the loop and spins narrow subagents for each phase.

Quality

94%

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 strong, highly actionable orchestration skill with a clear phased workflow, explicit validation gates and feedback loops, and concrete commands throughout. Its weakest spots are length and the inlining of detail that could be deferred to referenced files.

Suggestions

Tighten the 'Keep the orchestration console current' and 'Multi-agent coordination' sections — both run long; consider condensing the per-phase console hooks into a compact table and linking to a coordination doc for the serialization rules.

Move the QA matrix axes detail (E1–E4 environments, harness/surface definitions) into the already-referenced qa/matrix.md and keep only the 'force each capability with a token the model cannot guess' principle inline, reducing body length and improving progressive disclosure.

The Phase 4 matrix description repeats the environment/harness breakdown that lives in agent-workflows-qa; reference it once instead of restating, to trim conciseness without losing actionability.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and assumes Claude's competence — it gives repo-specific operational detail (GitButler gotchas, dev-mode reload rules, QA matrix axes) rather than explaining general concepts — but at ~248 lines sections like the orchestration-console hooks and multi-agent coordination could be trimmed; not 5 because not every token earns its place, not 3 because little is genuinely redundant.	4 / 5
Actionability	Provides copy-paste-ready commands ("but oplog snapshot -m ...", "but rub <path> <branch>", "but commit <branch> --only -m ...", "ruff check --fix", "gh pr create --head <branch> --base <parent-or-main>") and concrete file references (context.md, plan.md, qa/matrix.md), covering the common phases fully.	5 / 5
Workflow Clarity	Seven explicitly sequenced phases (0–6) with validation gates ("Do not advance with an open review", "Exit the loop when the slice works end to end", "Loop until the suite is green"), feedback loops (review→reimplement, debug run/observe/fix/re-run), and loop caps — matching the checklist-with-explicit-validation anchor.	5 / 5
Progressive Disclosure	Well structured with a phase diagram and a Quick reference table, and it clearly signals one-level references to sibling skills ("See debug-local-deployment for the exact commands", "Consult qa/matrix.md"); not 5 because there are no bundle files and detailed sections like the QA matrix axes and multi-agent coordination are inlined where a referenced file could carry them, not 3 because navigation is easy and references are clearly signaled.	4 / 5
	Total	18 / 20 Passed

Description

100%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.

An excellent description: concrete, explicitly states both what it does and when to use it with natural trigger phrases, and carves out a distinct niche as the implementation step after plan-feature. It is dense without being padded.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete orchestration actions — "refresh-plan, implement, review, a debug-local-deployment loop, and a test loop across the daytona / local-pi / claude x SDK / UI matrix, then documentation and a GitButler stacked branch" — covering the full pipeline comprehensively, matching the score-5 anchor rather than the several-actions-with-minor-gaps anchor at 4.	5 / 5
Completeness	Explicitly answers both what ("Drive a researched and planned feature to a landed, tested change") and when ("Use after plan-feature has produced a docs/design/<project>/ workspace and the user says...") with concrete trigger phrases, matching the score-5 anchor.	5 / 5
Trigger Term Quality	Includes four natural user phrases — "implement it", "build the plan", "run the plan", or "let's ship this" — which are exactly what a user would say, giving comprehensive synonym coverage; not 4 because no common variation is missing.	5 / 5
Distinctiveness Conflict Risk	Occupies a clear niche — the post-plan-feature implementation step — gated on a specific precondition and distinct trigger phrases, with minimal overlap risk against sibling skills like plan-feature or debug-local-deployment.	5 / 5
	Total	20 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning

	Total	15 / 16 Passed

Repository: Agenta-AI/agenta
Path: .agents/skills/implement-feature/SKILL.md
Commit: e4008fa

Reviewed: 5 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.