agentleFS
WrosinskiWrosinski/heddle/.claude/skills/new-feature/SKILL.md

new-feature

Skill0 starsChanged 57 days agoMITView on GitHub

Kick off a new feature through Phase 1 (Research & Brainstorming) and Phase 2 (Specification). Takes an optional feature brief path. Use when starting a new feature from scratch within the current session.

43 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use evidence-driven-testing, greploop-apps 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.
  • jaydubya818Skill

    evidence-driven-testing

    Capture observable runtime proof with video, screenshots, or measured output. Use when a change needs evidence beyond test reports.

    Also in 6 projects that use this

  • jaydubya818Skill

    greploop-apps

    Run greploop through @greptile-apps when the normal trigger rejects a large PR. Use only with the companion greploop skill.

    Also in 4 projects that use this

  • VercelSkill

    before-and-after

    Add existing screenshots or screen recordings to a GitHub pull request as a before/after or preview block. Use when a PR needs visual media attached to its description. Browser navigation and capture belong to agent-browser.

    Also in 6 projects that use this

  • BboyfaroukSkill

    greploop

    Iteratively improves a PR until Greptile gives it a 5/5 confidence score with zero unresolved comments. Triggers Greptile review, fixes all actionable comments, pushes, re-triggers review, and repeats. Use when the user wants to fully optimize a PR against Greptile's code review standards.

    Also in 6 projects that use this

  • u9401066Skill

    bug-fix

    Structured bug fixing workflow with root cause analysis and regression prevention. Triggers: BF, 修 bug, fix bug, debug, 除錯, 錯誤, error, 問題, issue, 修復, fix, 故障, fault, 異常, exception, 失敗, failure, 不能用, broken, 壞掉.

    Also in 3 projects that use this

  • SvampenSkill

    init-project

    Generates a tailored .claude/project-config.md for the current project. Asks about project type, quality check commands, documentation paths, insight tags, and review strictness.

    Also in 4 projects that use this

  • +1 more from jaydubya818

The file

We split this skill into the rules it gives an agent

11 rules. 0 are common, 3 are distinctive, and 8 are very specific to this project.

  • 0

    Common

    Most of the 41 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 8

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Read a supplied brief before entering the workflow.

    line 28
  2. Very specific:

    Run heddle orient --json. If it resolves existing work, follow its emitted routing and resume that feature.

    line 34
  3. Very specific:

    Only after formal policy confirmation, run heddle feature start <slug>.

    line 54
  4. Very specific:

    Follow returned next_actions, including a feature switch when emitted.

    line 55
  5. Very specific:

    Run heddle orient --feature <slug> --json, then heddle kickoff --feature <slug> only when orientation routes there.

    line 57
  6. Very specific:

    Use heddle/resources/decision-routing.md for ownership and question quality.

    line 62
  7. Distinctive:

    Use these sources for their respective contracts:

    line 75
  8. Distinctive:

    Use independent exploration only when available, authorized, and useful beside local work.

    line 85
  9. Very specific:

    Run heddle validate at the briefing's checkpoints.

    line 88
  10. Very specific:

    Never hand-edit state.yaml or intake records.

    line 94
  11. Distinctive:

    Keep one canonical document pair and one native progress owner.

    line 97

“Common” is measured: kept by at least 6 in 10 of the 41 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

New Feature Inception

Enter Heddle's native intake and, when selected, its feature workflow. This skill is a launcher, not a second policy resolver or copy of the workflow.

When to Use

  • Starting genuinely new feature work from a verbal request or written brief
  • Discovering related specifications and deciding the smallest useful scope
  • Producing the Feature Spec and Implementation Plan for a confirmed Heddle route

For an in-progress feature, resume the existing workspace instead of starting another one. For analysis without feature inception, use /pre-implementation-analysis.

Invocation

/new-feature [brief-path]

Read a supplied brief before entering the workflow. Keep the original research reference intact; formal admission binds its identity. After admission, enrich the generated plans/<slug>/brief.md without rewriting the approved provenance.

Native Entry Sequence

  1. Run heddle orient --json. If it resolves existing work, follow its emitted routing and resume that feature; do not create a duplicate workspace.
  2. For new work, research the intended outcome and actual integration surface under docs/workflow/brainstorming-guidelines.md. Retain a research reference and propose the route plus scope, complexity, and testability with reasons. Describe the concept in the brief's ## Concept section, offer the opt-in concept review, and record the owner's approval there at the research checkpoint, before step 3 binds the research reference. Record there too whether the owner skips the specification checkpoint or the review change confirmation for this feature (both on unless checkpoints in .heddle.yaml turns one off), and when the specification checkpoint will not run, the e2e execution grant.
  3. Use heddle feature prepare <slug> --area <area> --from-file <research-input> to record the intake. Native help defines its input contract. Preparation creates no formal feature, and its recommended matrix is not approval.
  4. Confirm the user's route. Direct work creates no feature workspace, spec, plan, or review schedule. For the Heddle route, present the complete resolved matrix and record the user's actual confirmation with heddle feature policy <slug> --from-file <confirmed-policy> --expect-revision <N>. Do not infer approval from the recommendation or rewrite host exclusions.
  5. Only after formal policy confirmation, run heddle feature start <slug>. Follow returned next_actions, including a feature switch when emitted; do not infer an alternate transition or introduce a tier argument.
  6. Run heddle orient --feature <slug> --json, then heddle kickoff --feature <slug> only when orientation routes there.

Decision ownership

Use heddle/resources/decision-routing.md for ownership and question quality; feature kickoff delivers the resolved host policy in full. Preserve this skill's analysis/authoring boundary and applicable standing authorization. A review or diagnosis request alone grants no implementation authority.

Research and Specification Contract

For an admitted feature, kickoff and the packaged specify briefing own the current stage. The research checkpoint is distinct from the specification checkpoint at specify exit and from the review change confirmation after each document review. Honor already-recorded research and scope choices; reopen them only for material new evidence. This skill adds no separate approval step.

Use these sources for their respective contracts:

  • docs/workflow/brainstorming-guidelines.md: research, decomposition, consequential commitments, and per-part scope authorization;
  • docs/workflow/workflow.md: confirmation, execution authority, and lifecycle;
  • docs/workflow/document-structure.md: spec commitments and canonical ACs, plan approach and technical architecture, and the knowledge index; and
  • heddle search "<intent>" --titles-only: related specs and precedents before broad directory scanning.

Use independent exploration only when available, authorized, and useful beside local work. Synthesize its evidence into one research result.

Run heddle validate at the briefing's checkpoints. Advance only through heddle phase-exit or the driver when the runtime and execution authority permit it. Required decisions and authorization stops remain mandatory.

Constraints

  • Never hand-edit state.yaml or intake records; native commands own them.
  • For a Heddle route, this skill authors research and specifications, not production implementation. Direct work follows the user's scoped request.
  • Keep one canonical document pair and one native progress owner.
  • On a native refusal, follow its exact diagnostic and legal remedy; do not substitute a remembered legacy procedure or silently fall back to a provider.

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.