agentleFS
naman10parikhnaman10parikh/pixel/.claude/skills/repo-scout/SKILL.md

repo-scout

Skill0 starsChanged 28 days agoMITView on GitHub

Scout popular GitHub repos for patterns, frameworks, and harnesses relevant to Energy. Ingest learnings automatically.

8 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 2 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 2

    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:

    When discovering a new framework/tool and wanting to find usage examples

    line 12
  2. Very specific:

    Write findings to resources/read/repo-scout-{date}.md:

    line 55
  3. Very specific:

    Update resources/awesome-repos.md with new entries

    line 71
  4. Very specific:

    If a pattern is novel, create a learning in memory/LEARNINGS.md

    line 72
  5. Distinctive:

    If a framework is new to us, add to MEMORY.md tech choices

    line 73

“Common” is measured: kept by at least 6 in 10 of the 4 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

Repo Scout — Continuous Learning from the Ecosystem

Trigger

  • User says "scout repos", "find repos", "what's new on GitHub"
  • Periodic execution via GHA or session start
  • When discovering a new framework/tool and wanting to find usage examples

Process

1. Discovery (parallel sub-agents)

For each search query, launch a background agent:

gh search repos --sort=stars --limit=10 "QUERY"

Default search queries (expand as stack evolves):

  • e2b sandbox agent — sandbox-based agents
  • claude agent sdk — Claude Agent SDK usage
  • mcp server tools — MCP integrations
  • ai agent browser automation — browser agents
  • composio agent — Composio-based agents
  • vercel ai sdk chat — AI chat UIs
  • assistant-ui chat — assistant-ui implementations
  • shadcn agent — shadcn/ui + agent combos
  • stagehand browser — Stagehand browser automation
  • copilotkit agent — CopilotKit implementations

2. Analysis (per repo)

For each promising repo (>50 stars, recent activity):

gh api repos/{owner}/{name}/git/trees/main?recursive=1

Extract:

  • Architecture: File structure, package.json deps, framework choices
  • Agent patterns: How they handle tool calls, state machines, error recovery
  • UI patterns: Status updates, split-screen, generative UI
  • Browser patterns: E2B, Playwright, Stagehand, computer use
  • Harness patterns: Skills, memory, context management

3. Document

Write findings to resources/read/repo-scout-{date}.md:

# Repo Scout — {date}

## {repo-name} ({stars} stars)

- **URL**: {url}
- **Stack**: {frameworks used}
- **Key pattern**: {one-line insight}
- **Files to study**: {list of key files}
- **Adoptable for Energy**: {what we can use}

4. Integrate

  • Update resources/awesome-repos.md with new entries
  • If a pattern is novel, create a learning in memory/LEARNINGS.md
  • If a framework is new to us, add to MEMORY.md tech choices

Reference Repos (Permanent Watchlist)

RepoWhy
ComposioHQ/open-gensparkOpen-source GenSpark clone using Composio
FoundationAgents/OpenManusOpen-source Manus clone
anthropics/claude-codeOfficial Claude Code patterns + plugins
AbanteAI/nanoclawMinimal Claude Agent SDK reference
e2b-dev/fragmentsE2B + AI code generation
mckaywrigley/chatbot-uiReference chat UI
stackblitz/bolt.newBrowser-based AI builder
all-hands-ai/openhandsOpen-source Devin
browser-use/browser-useBrowser automation for AI
browserbase/stagehandTS-native browser control
hesreallyhim/awesome-claude-codeMaster Claude Code ecosystem index
rohitg00/awesome-claude-code-toolkit135 agents, 42 commands, 7 templates
affaan-m/everything-claude-codeHackathon winner: 65+ skills, 40 commands
wshobson/commands57 production slash commands
trailofbits/claude-code-configSecurity-focused Claude Code config
VoltAgent/awesome-claude-code-subagents100+ sub-agent definitions
punkpeye/awesome-mcp-servers2,500+ MCP servers catalog

Output

  • resources/read/repo-scout-{date}.md — findings document
  • Updated resources/awesome-repos.md — catalog
  • New learnings in memory/LEARNINGS.md if applicable

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.