本文へ移動
cccskills
無料GitHub で公開

external-pattern-integrator

Integrates patterns from external sources (URLs or local files) into local skills, agents, and plugins. Use when comparing external agent definitions against local equivalents, extracting best practices from frameworks like GSD or BMAD-METHOD, enhancing local skills with external workflow patterns, or adding interoperability with external tool ecosystems. Runs a 3-phase workflow — parallel candidate mapping, contextual enhancement, and validation — with source attribution and backlog tracking for deferred enhancements.

インストール方法を見る

含まれるファイル(1)

  • SKILL.md8.7 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

<source_refs>$ARGUMENTS</source_refs>

External Pattern Integrator

Systematically analyze external agent/skill definitions and integrate their strengths into local skills while maintaining workflow coherence.

Arguments

<source_refs/>

Workflow Overview

Phase 1: Parallel Candidate Mapping
├── For each external source (concurrent):
│   ├── Fetch/read the external file
│   ├── Analyze its purpose and patterns
│   └── Scan local files for candidates (frontmatter only)
└── Output: Tracking document with source → candidates mapping

Phase 2: Contextual Enhancement (concurrent per source)
├── Understand local file's workflow context FIRST
├── Compare against external patterns
├── Identify enhancements that fit the workflow stage
├── Add external artifact recognition for interoperability
└── Edit files with coordinated changes

Phase 3: Validation
└── Run linting on all modified files

Phase 1: Parallel Candidate Mapping

Step 1.1: Create Tracking Document

Create a tracking file at .claude/external-pattern-integration-{date}.md:

# External Pattern Integration

**Date**: {YYYY-MM-DD}
**Sources**: {list of URLs/files from arguments}
**Status**: IN_PROGRESS

## Source Analysis

### Source 1: {URL or path}

**Purpose**: {What this external agent/skill does}
**Key Patterns**:
- {Pattern 1}
- {Pattern 2}

**Local Candidates** (by frontmatter similarity):
| Local File | Similarity Reason | Priority |
|------------|-------------------|----------|
| {path} | {why it matches} | {High/Medium/Low} |

### Source 2: {URL or path}
...

Step 1.2: Fetch External Sources

For each URL/file in <source_refs/>:

If URL: Use WebFetch or curl to download to /tmp/external-pattern-{slug}.md If local file: Read directly

Step 1.3: Analyze External Source

For each external source, extract:

  1. Purpose: What problem does it solve?
  2. Key Patterns: What techniques does it use?
  3. Artifact Files: What files does it create/read? (for interoperability)
  4. Workflow Stage: Where in a development workflow does it fit?

Step 1.4: Scan Local Candidates

Scan these locations for candidates (frontmatter only, not full files):

plugins/*/skills/*/SKILL.md
plugins/*/agents/*.md
.claude/skills/*/SKILL.md
.claude/agents/*.md

Match by:

  • Similar purpose (from description field)
  • Similar workflow stage
  • Overlapping functionality

Priority Assignment:

  • High: Direct functional overlap (same workflow stage, similar purpose)
  • Medium: Adjacent functionality (related workflow stage)
  • Low: Tangential (shared patterns but different purpose)

Step 1.5: Deduplicate Candidates

If a local file appears as candidate for multiple external sources:

  1. Assign to the BEST match (highest similarity)
  2. Note secondary matches for cross-reference during enhancement

Phase 2: Contextual Enhancement

Run concurrently for each external source. For each candidate local file:

Step 2.1: Understand Workflow Context

BEFORE comparing, determine:

  1. Is this file part of a multi-file workflow?

    • Check for references to other skills/agents in the file
    • Check if other files reference this one
    • Look for phase indicators (discovery → planning → implementation → verification)
  2. What stage of the workflow is this file for?

    • Discovery/Research (gathering context)
    • Planning (creating specs, acceptance criteria)
    • Implementation (writing code)
    • Verification (testing, validation)
    • Orchestration (coordinating other agents)
  3. What are the upstream/downstream dependencies?

    • What artifacts does this file expect as input?
    • What artifacts does this file produce?

Step 2.2: Compare and Identify Enhancements

Read the full external source and full local file. Identify:

Enhancement TypeDescriptionFits Current Stage?
Missing patternExternal has X, local lacks itYes/No
Stronger guidanceExternal has better instructions for YYes/No
Better structureExternal organizes Z more clearlyYes/No
Artifact recognitionExternal creates files local should recognizeYes/No

Critical Rule: Only recommend enhancements that fit the file's workflow stage. If an enhancement belongs in a different stage, note which file should receive it instead.

Step 2.3: Add External Artifact Recognition

For interoperability, add a section to context-gathering skills:

## External Framework Artifacts

<external_artifacts>

When gathering context, also check for these artifacts from external frameworks:

**Get Shit Done (GSD)**:
- `STATE.md` - Current project state and progress
- `ROADMAP.md` - Feature roadmap and planning
- `codebase-map.md` - Generated codebase structure
- `research-*.md` - Research documents
- `plan-*.md` - Execution plans

**BMAD-METHOD**:
- `*.agent.yaml` - Agent definitions
- `workflows/*.md` - Workflow definitions
- `party-mode-session.md` - Multi-agent collaboration notes

If found, incorporate their context into discovery.

</external_artifacts>

Step 2.4: Coordinate Cross-File Changes

If enhancements affect multiple files in a workflow:

  1. List all affected files
  2. Determine the order of changes (upstream before downstream)
  3. Ensure consistency (same terminology, compatible artifacts)
  4. Make changes atomically (all or none)

Step 2.5: Apply Enhancements

Edit local files with:

  • Clear section markers for new content
  • Source attribution: SOURCE: Adapted from {external source URL/path}
  • Preserved existing functionality (additive changes preferred)

Phase 3: Validation

Step 3.1: Run Linting

uv run prek run --files {all modified files}

Fix any issues before proceeding.

Step 3.2: Update Tracking Document

Update the tracking file with:

## Results

**Files Modified**:
| File | Enhancements Applied | Source |
|------|---------------------|--------|
| {path} | {what was added} | {external source} |

**Deferred Enhancements** (didn't fit current files):
| Enhancement | Reason Deferred | Suggested Location |
|-------------|-----------------|-------------------|
| {pattern} | {why} | {where it should go} |

**Status**: COMPLETE

Step 3.3: Add Deferred Items to Backlog

If there are deferred enhancements, create per-item files in .claude/backlog/ via the backlog script:

  1. For each deferred enhancement, invoke the backlog script with appropriate priority:
    • P1 (Should Have): Patterns that would significantly improve workflows
    • P2 (Could Have): Nice-to-have patterns or minor improvements
    • Ideas: Patterns worth exploring but unclear fit

Entry format:

### {Enhancement title}

**Source**: [{tracking-document}]({path-to-tracking-document})
**Added**: {YYYY-MM-DD}
**Description**: {What needs to be done}
**Patterns from**: {external-source-name}
**Suggested location**: {path/to/file.md}
  1. Note in tracking document: "Deferred items added to .claude/backlog/ as per-item files"

Step 3.4: Commit Changes

Stage and commit with message:

feat(skills): integrate patterns from {source names}

Enhancements:
- {file1}: {what was added}
- {file2}: {what was added}

Sources:
- {URL1}
- {URL2}

Success Criteria

  • All external sources fetched and analyzed
  • Tracking document created with candidate mapping
  • Workflow context understood before making changes
  • Enhancements fit the file's workflow stage
  • Cross-file changes coordinated
  • External artifact recognition added for interoperability
  • All modified files pass linting
  • Deferred enhancements added as per-item files in .claude/backlog/ with priority
  • Changes committed with source attribution

Example Usage

User: /external-pattern-integrator https://github.com/glittercowboy/get-shit-done/blob/main/agents/gsd-codebase-mapper.md

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Add automated documentation updater to any Claude skill. Creates a Python sync script that downloads upstream docs, processes markdown for AI consumption, and maintains local cache with configurable refresh. Collects template variables, then delegates implementation through 5-phase workflow. Use when adding auto-updating reference documentation to plugins or skills.

日本語の概要は準備中です。原文の説明を表示しています。

Jamie-BitFlight/claude_skills672026年10月9日 更新

SAM-style feature initiation workflow — discovery through codebase analysis, architecture spec, task decomposition, validation, and context manifest. Use when a user asks to add a feature, plan a feature, or convert an idea into an executable SAM plan.

日本語の概要は準備中です。原文の説明を表示しています。

Jamie-BitFlight/claude_skills672026年10月9日 更新

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction.

日本語の概要は準備中です。原文の説明を表示しています。

Jamie-BitFlight/claude_skills672026年10月9日 更新

Browser automation for AI agents using the agent-browser CLI and Playwright. Use when navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, logging into sites, or automating any browser task. Triggers on "open a website", "fill out a form", "click a button", "scrape data", "test this web app", "automate browser actions", or any programmatic web interaction request.

日本語の概要は準備中です。原文の説明を表示しています。

Jamie-BitFlight/claude_skills672026年10月9日 更新

Runs the description-drift experiment — spawns all Claude Code agents simultaneously to collect self-reported capabilities, then compares them against static frontmatter descriptions to reveal how reliable orchestrator routing based on descriptions actually is. Use when measuring description drift across the agent fleet, re-running the capability collection experiment, analyzing a specific agent's self-reported capabilities, or auditing whether frontmatter descriptions accurately reflect agent behavior.

日本語の概要は準備中です。原文の説明を表示しています。

Jamie-BitFlight/claude_skills672026年10月9日 更新

Create or adapt Claude Code agent definitions. Use when creating an agent, changing subagent configuration, selecting agent scope, or designing a specialized delegation role.

日本語の概要は準備中です。原文の説明を表示しています。

Jamie-BitFlight/claude_skills672026年10月9日 更新

Jamie-BitFlight のスキルをすべて見る

このスキルの問題を報告する