swingerman/engineer/feature-edit
Claude Code and agent skill by swingerman/engineer
- community
Checking that the SKILL.md file is still in the repository.
Use when an in-flight feature must change — scope shifts, an AC is refined, the plan evolves, a behavior is added or removed. Triggers — "/engineer.feature-edit", "change feature X to", "the ACs need updating", "revise the plan", "scope changed".
Install
npx vskill@latest install swingerman/engineer/feature-editSKILL.md
feature-edit
Intent-driven editing of an in-flight feature. The human describes what should now be true; this skill figures out which files and tools to touch and orchestrates the change, including the downstream cascade.
The cascade-owning skill — clarify and refine deliberately stay single-artifact and route multi-artifact propagation here. Mixed mode: the agent proposes an edit plan, the human confirms before any write.
When to use
Any time an existing feature's artifacts must change. checkpoint: null.
Not for: a brand-new capability (discuss); within-one-artifact ambiguity, no behavior change (clarify); validating consistency (consistency-check); cleaning up code (refine).
Workflow
- Resolve + identify the feature — resolve the methodology root + manifest via
${CLAUDE_PLUGIN_ROOT}/scripts/dae_resolve.py(seereferences/resolving.md); locate the feature (slug, branch, or searchfeature.mdtitles/outcomes). - Is this actually an edit? — if the intent is a new capability or too big for one feature, surface it and redirect to
discuss. Stop. - Classify the entry artifact — the highest-level artifact the intent directly changes: outcome/scope →
feature.md; a behavior →acs.md; a GWT detail →spec.md; architecture/phasing →plan.md. - Build the edit plan — the cascade runs strictly downstream (
feature.md → acs.md → spec.md → plan.md); upstream problems are surfaced, never auto-propagated. Per downstream artifact, classify by blast radius: small/mechanical → feature-edit edits directly; substantial → invoke the owning skill (discover-acs/atdd:atdd/plan) in edit-pass mode. - Present the plan; human confirms — show entry artifact, cascade, and how each step is handled. Nothing is written before confirmation.
- Execute — in cascade order: direct edits and owning-skill invocations; regenerate
.build/spec.jsonafter anyspec.mdchange; re-run affected test streams if code exists. If regeneration reveals an upstream problem, stop and surface it. - Sync — update
progress.md(affected checkpoint rows) and the tracker via the driver. If the edit adds or changes a user-facing surface (route, command, screen), update that feature's row in theverify-<app>feature map (${CLAUDE_PLUGIN_ROOT}/references/verify-app-skill.md). - Handoff — emit a summary.
Handoff
Emit per ${CLAUDE_PLUGIN_ROOT}/references/handoff-summary.md. checkpoint: null; artifacts lists every artifact edited. recommended_next: typically "re-run /engineer.consistency-check".
References
- Foundation Design — artifact schemas, the IR pipeline (Section 7)
- Discuss & Upstream Funnel — when an "edit" is really a new feature
Security
- Security scanPassed, 100 of 100Pattern checks on the skill's files.
- Model reviewNot runA model review is scheduled for the next rescan.
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingengineer/skills/feature-edit/SKILL.md
Popularity
- 7-day trend score
- 0
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/swingerman/engineer/feature-edit)