mcpbeat Sign in

Parent Project Skills

Bootstrap skill for discovering additional skills and context from a parent project when workerd is used as a submodule. Load this skill when tasks span project boundaries (e.g., Sentry/production investigation, integration testing, cross-repo debugging).

731 tokens
context cost
the whole folder, loaded on every use
1
files
instructions only
0
copies elsewhere
how many repositories repackaged it
8447
stars on the repo
on the repository, not the skill itself

Install

one command, takes just this skill from the repository
npx skills add https://github.com/cloudflare/workerd --skill parent-project-skills

What it tells the agent to use

found in the instruction text
Read reads your files

The instruction itself

4 sections, as written by the author

Parent Project Skill Discovery

This project (workerd) may be embedded as a submodule within a larger project. When your task requires context beyond workerd itself — such as investigating production issues, writing integration tests, or understanding deployment architecture — you should check for additional skills and context in the parent project.

Discovery procedure

  • Determine if a parent project exists by checking for a git root above the workerd directory:
  • Look for ../../.git or ../../AGENTS.md relative to the workerd root
  • If found, the parent project root is ../../ (two levels up from workerd)
  • Check for parent project skills:
  • Look for ../../.opencode/skills/*/SKILL.md
  • Read any ../../AGENTS.md or ../../AGENTS.md for project-wide context
  • Load relevant skills by reading the SKILL.md files. These are not registered in the skill tool — read them directly with the Read tool.
  • Check for submodule-specific context:
  • Look for ../../.opencode/skills/*/SKILL.md files that may reference workerd specifically

Information boundary — CRITICAL

workerd is an open-source project. The parent project is typically proprietary/internal.

You MUST enforce a strict one-way information boundary:

  • ALLOWED: Internal context informs your reasoning, helps you understand code paths, guides your investigation
  • NEVER: Write internal details into workerd files — this includes:
  • Code comments referencing internal systems, services, or architecture
  • Commit messages mentioning internal projects, Sentry issues, Jira tickets, or internal URLs
  • PR descriptions or GitHub issue comments containing internal context
  • AGENTS.md, AGENTS.md, or documentation updates with internal knowledge
  • Variable names, error messages, or log strings that reveal internal details
  • Test files that encode internal architecture assumptions

When working across the boundary, frame all workerd-side artifacts in terms of workerd's own public concepts (Workers, Durable Objects, isolates, IoContext, etc.) without referencing how the parent project orchestrates them.

When to use this skill

  • Investigating production Sentry issues that involve workerd code
  • Writing integration tests (e.g., ew-test) that live in the parent project
  • Understanding how workerd APIs behave in the production deployment context
  • Debugging issues that cross the workerd / parent-project boundary
  • Reviewing code changes that affect both projects
  • Planning new feature development that requires coordination between workerd and the parent project

Other skills for the same job

different authors, same section of the catalogue
Loki Mode
by ComeOnOliver
×2

Multi-agent autonomous startup system for Claude Code. Triggers on "Loki Mode". Orchestrates 100+ specialized agents across engineering, QA, DevOps, security, data/ML, business operations, marketing, HR, and customer success. Takes PRD to fully deployed, revenue-generating product with zero human intervention. Features Task tool for subagent dispatch, parallel code review with 3 specialized reviewers, severity-based issue triage, distributed task queue with dead letter handling, automatic deployment to cloud providers, A/B testing, customer feedback loops, incident response, circuit breakers, and self-healing. Handles rate limits via distributed state checkpoints and auto-resume with exponential backoff. Requires --dangerously-skip-permissions flag.

3528k tokens scripts
Error Debugging Multi Agent Review
by lingxling
×1

Use when working with error debugging multi agent review

2k tokens
Evaluation
by lingxling
×1

Build evaluation frameworks for agent systems. Use when testing agent performance systematically, validating context engineering choices, or measuring improvements over time.

3k tokens
Agent Communication Debugger
by ComeOnOliver
×1

Diagnoses and debugs A2A agent communication issues including agent status, message routing, transport connectivity, and log analysis. Use when agents aren't responding, messages aren't being delivered, routing is incorrect, or when debugging orchestrator, coder-agent, tester-agent communication problems.

23k tokens scripts
Error Debugging Multi Agent Review
by ComeOnOliver
×1

Use when working with error debugging multi agent review

4k tokens
Micro Skill Creator
by ComeOnOliver
×1

Rapidly creates atomic, focused skills optimized with evidence-based prompting, specialist agents, and systematic testing. Each micro-skill does one thing exceptionally well using self-consistency, program-of-thought, and plan-and-solve patterns. Enhanced with agent-creator principles and functionality-audit validation. Perfect for building composable workflow components.

6k tokens
Ohmg
by ComeOnOliver
×1

Ultimate multi-agent framework for Google Antigravity. Orchestrates specialized domain agents (PM, Frontend, Backend, Mobile, QA, Debug) via Serena Memory.

3k tokens
Evaluation
by ComeOnOliver
×1

This skill should be used when the user asks to "evaluate agent performance", "build test framework", "measure agent quality", "create evaluation rubrics", or mentions LLM-as-judge, multi-dimensional evaluation, agent testing, or quality gates for agent pipelines.

13k tokens scripts

How to use it

Copy the folder

Take cloudflare/parent-project-skills from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.