agentleFS
HorizonRoboticsHorizonRobotics/RoboOrchardLab/.agents/skills/feature-dev/SKILL.md

feature-dev

Skill180 starsChanged 9 days agoApache-2.0View on GitHub

Guided feature development with codebase understanding and architecture focus. Use when implementing a non-trivial feature that benefits from structured discovery, exploration, architecture comparison, implementation, and review.

21 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use code-review.

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.
  • GoogleSkill

    code-review

    Adversarial systems, concurrency, memory-model, and architecture code review protocol. Use this skill whenever asked to review code, PRs, diffs, or branches in Filament.

    Also in 6 projects that use this

The file

We split this skill into the rules it gives an agent

12 rules. 0 are common, 9 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 17 edited versions we read keep these.

  • 9

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill to coordinate non-trivial feature work from discovery through implementation, review, validation, and summary.

    line 6
  2. Distinctive:

    Before running this workflow, read:

    line 11
  3. Distinctive:

    Use the sub-skills in this directory only when their phase applies:

    line 15
  4. Distinctive:

    Ask concrete clarifying questions when behavior, scope, compatibility, or integration points are underspecified.

    line 30
  5. Distinctive:

    Read files surfaced by exploration or design passes before making architectural claims.

    line 32
  6. Distinctive:

    Prefer simple, maintainable designs grounded in repository patterns.

    line 34
  7. Very specific:

    Use code-explorer for broad or unfamiliar areas before designing.

    line 54
  8. Very specific:

    Use code-architect when multiple credible architecture choices exist.

    line 55
  9. Very specific:

    Use code-reviewer for non-trivial implementation review or explicit review loops.

    line 56
  10. Distinctive:

    Default to at most one active delegated reviewer per review round.

    line 58
  11. Distinctive:

    Use parallel reviewers within one round only when the user explicitly requests them or a concrete high-risk scope justifies independent …

    line 64
  12. Distinctive:

    After delegated review, the main agent still owns consolidation, validation, and final handoff quality.

    line 66

“Common” is measured: kept by at least 6 in 10 of the 17 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

Use this skill to coordinate non-trivial feature work from discovery through implementation, review, validation, and summary.

Read Next

Before running this workflow, read:

  • references/phase-workflow.md

Use the sub-skills in this directory only when their phase applies:

  • code-explorer: deep codebase tracing and implementation analysis
  • code-architect: architecture design and implementation blueprinting
  • code-reviewer: high-signal review for correctness and project fit

Role

This parent skill is the orchestration layer. Keep it focused on routing, phase selection, and handoff between sub-skills. Concrete execution details belong in the phase workflow reference or in the smallest applicable sub-skill.

Core Principles

  • Understand the codebase before changing it.
  • Ask concrete clarifying questions when behavior, scope, compatibility, or integration points are underspecified.
  • Read files surfaced by exploration or design passes before making architectural claims.
  • Prefer simple, maintainable designs grounded in repository patterns.
  • Track progress with a todo list when the work spans multiple steps.
  • Compress the workflow for trivial tasks, but preserve the same reasoning discipline.

Workflow Summary

The detailed actions, examples, and review rules live in references/phase-workflow.md. The phase order is:

  1. Discovery: understand the request and constraints.
  2. Codebase exploration: map relevant implementation patterns.
  3. Clarifying questions: resolve material ambiguity before designing.
  4. Architecture design: compare viable approaches and choose one.
  5. Implementation: build the approved or obvious approach.
  6. Quality review: run risk-appropriate review, validation, and Cleanup Gate.
  7. Summary: report decisions, changed files, validation, and residual risks.

Delegation Rules

  • Use code-explorer for broad or unfamiliar areas before designing.
  • Use code-architect when multiple credible architecture choices exist.
  • Use code-reviewer for non-trivial implementation review or explicit review loops.
  • Default to at most one active delegated reviewer per review round. Treat architecture, correctness, simplicity, and project conventions as review dimensions that one reviewer can cover together.
  • In a review loop, keep the current reviewer only long enough to verify fixes for its own findings, then start the next full-scope round with a fresh reviewer.
  • Use parallel reviewers within one round only when the user explicitly requests them or a concrete high-risk scope justifies independent review.
  • After delegated review, the main agent still owns consolidation, validation, and final handoff quality.

When Not To Use

This workflow is usually overkill for:

  • one-line bug fixes
  • trivial refactors
  • clearly scoped edits in a single file
  • urgent hotfixes where architecture comparison adds no value

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.