Skip to content

aspiers/ai-config / .agents/skills/project-initialization

project-initialization

Verified

Establish concise, cross-agent project instructions from repository evidence. Use when initializing or repairing AGENTS.md, CLAUDE.md, related discovery links, or project-specific agent guidance.

Project Initialization

Create a lightweight entry point that explains the repository and highlights non-obvious gotchas. Do not turn it into a catalog of generic development advice.

Workflow

  1. Inspect existing agent instructions, repository documentation, build and test interfaces, and file-discovery conventions supported by the target harnesses.
  2. Preserve useful existing guidance and identify conflicts before editing.
  3. Choose one authoritative instruction file when the repository wants shared content. Use symlinks or small platform-specific entry points only when they are supported by the repository and tools; do not rename files or replace independent guidance without agreement.
  4. Keep the entry point concise:
    • what the repository is;
    • authoritative build, test, and validation commands;
    • architecture or workflow facts that are not obvious from the tree; and
    • narrow safety or contribution constraints.
  5. Move detailed or conditional workflows into focused skills or directly linked documentation.
  6. Verify every discovery path, symlink, command, and cross-reference.

Safety review

Review generated instructions for commands that could publish, delete, overwrite, migrate, or expose data. Prefer explicit targets and narrow scope, but preserve intentional repository workflows rather than rewriting commands from a universal template.

Do not add secrets, personal conventions, or private operational details to a public repository. Avoid duplicating global harness rules or facts the model can obtain by inspecting the project.

Result

Report the authoritative file, compatibility links or wrappers, validation performed, and any platform whose discovery behavior remains uncertain.

reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 465kUpdated Claude CodeCodexVerifiedInstall
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 465kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 318kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 318kUpdated Claude CodeCodexVerifiedInstall