agentleFS
ianallowayianalloway/openclaw-skills/devin-integration/SKILL.md

devin-integration

Skill4 starsChanged 5 months agoMITView on GitHub

Delegate coding tasks to Devin AI agent. Create PRs, fix bugs, build features, and manage GitHub repos through Devin's API.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Keep in OpenClaw when:

    line 26
  2. Distinctive:

    When Ian asks you to do coding work, compose a clear task description:

    line 43
  3. Distinctive:

    All repos have AGENTS.md files that tell Devin how to work with them:

    line 88
  4. Distinctive:

    After Devin creates a PR:

    line 130
  5. Distinctive:

    Leave comments directly on the PR for changes

    line 133

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Devin Integration

Delegate software engineering tasks to Devin, an AI coding agent by Cognition AI. Devin can create PRs, fix bugs, add features, refactor code, and manage your GitHub repositories autonomously.

When to Use Devin

Delegate to Devin when:

  • Code changes needed: Bug fixes, new features, refactoring, dependency updates
  • PR creation: Any task that results in a pull request
  • Multi-file changes: Tasks spanning multiple files or repos
  • CI/CD work: Fixing build failures, adding tests, updating pipelines
  • Repo maintenance: Adding configs, updating docs, dependency management
  • Complex coding: Architecture changes, new integrations, API work

Keep in OpenClaw when:

  • Quick questions or research (use Perplexity)
  • Conversation, planning, brainstorming
  • Non-code tasks (email, calendar, reminders)
  • Real-time monitoring or alerts

How to Delegate to Devin

Via Devin Web App

Tell Ian to go to https://app.devin.ai and start a session with a prompt like:

Work on repo ianalloway/{repo-name}:
{description of what needs to be done}

Composing Tasks for Devin

When Ian asks you to do coding work, compose a clear task description:

Repo: ianalloway/{repo-name}
Branch: main (or specify)
Task: {clear description}
Files to focus on: {if known}
Acceptance criteria:
- {criterion 1}
- {criterion 2}
Test command: {npm run test / pytest / etc}
Lint command: {npm run lint / etc}

Example Delegations

Fix a bug:

Repo: ianalloway/ian-web-forge
Task: The donation button on the portfolio site links to the wrong ETH address.
Fix: Update to 0xAc7C093B312700614C80Ba3e0509f8dEde03515b
Test: npm run build should pass

Add a feature:

Repo: ianalloway/sports-betting-ml
Task: Add NBA player prop predictions to the Streamlit dashboard.
Requirements:
- New tab in the dashboard for player props
- Use existing model architecture
- Add unit tests
Test: pytest

Cross-repo update:

Update all three locations for Ian's new project:
1. ianalloway/Resume - Add project to CV
2. ianalloway/ianalloway - Update GitHub profile README
3. ianalloway/ian-web-forge - Add to portfolio projects section

Ian's Repositories (Devin-Ready)

All repos have AGENTS.md files that tell Devin how to work with them:

Web/Frontend (React + Vite + TypeScript + shadcn/ui)

RepoPurposeCommands
ian-web-forgePortfolio website (ianalloway.xyz)npm run dev/build/lint
ai-advantageSports betting AI platformnpm run dev/build/lint
mayc-ai-assist-23MAYC NFT AI assistantnpm run dev/build/lint
fairmont-alloway-estimateBusiness estimation toolnpm run dev/build/lint
paint-prep-list-makerPaint prep checklist toolnpm run dev/build/lint
kana-dojoJapanese learning platformnpm run dev/build/lint/test

Python/ML

RepoPurposeCommands
ResumeCareer agent + CV toolspython main.py
sports-betting-mlML betting models + Streamlitpython app.py, pytest
ai-drone-auto-vehicleAutonomous vehicle/drone AIDocker-based
crypto-portfolio-cliCrypto portfolio trackerpython portfolio.py, pytest
ai-portfolio-analyzerInvestment portfolio analysispython portfolio_analyzer.py

OpenClaw Ecosystem

RepoPurpose
Money-maker-botOpenClaw core (Clawdbot fork)
openclaw-skillsCustom skills for OpenClaw
openclaw-brainBooper Bot system files (AGENTS.md, SOUL.md, etc.)
open-clawOpenClaw project hub
clawhubSkill directory for OpenClaw
booper-brainBooper Bot personal notes
booper-projectsJob tracker + project ideas

Other

RepoPurpose
ianallowayGitHub profile README
Mutant-IntelNFT AI agent
PeekaboomacOS screenshot CLI + MCP server
R-programming-assignmentUSF coursework

Checking Devin's Work

After Devin creates a PR:

  1. Ian will get a notification with the PR link
  2. Review the PR on GitHub
  3. Leave comments directly on the PR for changes
  4. Devin monitors and responds to PR comments
  5. Merge when satisfied

Tips

  • Devin works best with clear, specific task descriptions
  • Include the repo name and any relevant file paths
  • Mention test/lint commands so Devin can verify its work
  • For cross-repo tasks, list all repos that need updating
  • Devin creates PRs (never pushes directly to main)

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.