agentsclimarketplace

Specs explore

Skill giuseppe-trisciuoglio/developer-kit/plugins/developer-kit-specs/skills/specs-explore

Explore codebase before committing to a change. Phase executor skill for specs.explore command.From its SKILL.md

Install
npx -y skills add giuseppe-trisciuoglio/developer-kit --skill specs-explore

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its file declares

Copied from the file, not written here

The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

3.0 KB, 736 tokens by cl100k_base, as published. Nobody here has run it

specs-explore Skill

Investigate the codebase, think through problems, compare approaches, and return a structured analysis. Only create exploration.md when --spec is provided.

What You Receive

The orchestrator will give you:

  • A topic or feature to explore
  • --spec path (optional)
  • --lang framework (optional)

Execution Steps

Step 1: Understand the Request

Parse what the user wants to explore:

  • Is this a new feature? A bug fix? A refactor?
  • What domain does it touch?

Step 2: Investigate the Codebase

Read relevant code to understand:

  • Current architecture and patterns
  • Files and modules that would be affected
  • Existing behavior that relates to the request
  • Potential constraints or risks
INVESTIGATE:
├── Read entry points and key files
├── Search for related functionality (Grep)
├── Check existing tests (Glob)
├── Look for patterns already in use
└── Identify dependencies and coupling

Step 3: Analyze Options

If there are multiple approaches, compare them:

ApproachProsConsComplexity
Option A......Low/Med/High
Option B......Low/Med/High

Step 4: Persist Artifact (if --spec is provided)

If --spec=docs/specs/001-add-jwt-auth is specified, write to docs/specs/001-add-jwt-auth/exploration.md:

## Exploration: {topic}

### Current State
{How the system works today relevant to this topic}

### Affected Areas
- `path/to/file.ext` — {why it's affected}
- `path/to/other.ext` — {why it's affected}

### Approaches
1. **{Approach name}** — {brief description}
   - Pros: {list}
   - Cons: {list}
   - Effort: {Low/Medium/High}

2. **{Approach name}** — {brief description}
   - Pros: {list}
   - Cons: {list}
   - Effort: {Low/Medium/High}

### Recommendation
{Your recommended approach and why}

### Risks
- {Risk 1}
- {Risk 2}

### Ready for Proposal
{Yes/No — and what the user should know}

Step 5: Return Result

Return the same structured analysis (inline), plus:

Status Report:

  • status: done | blocked | partial
  • executive_summary: 1-sentence description of what was explored and key recommendation
  • artifact: {spec}/exploration.md (if --spec provided) or "none"
  • next_recommended: specs.brainstorm (if spec provided) or none (if standalone)
  • risks: risks discovered, or "None"

Rules

  • The ONLY file you MAY create is exploration.md inside the spec folder (if --spec provided)
  • DO NOT modify any existing code or files
  • ALWAYS read real code, never guess about the codebase
  • Keep your analysis CONCISE — summary, not a novel
  • If you can't find enough information, say so clearly
  • If the request is too vague, ask for clarification

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.