repo-memory
Repository Memory
Use this skill to keep a small, durable memory inside the repository so future work starts from the current technical and human context instead of rebuilding it from scratch.
If the user explicitly invokes this skill, treat that as permission to bootstrap the repository memory system in the current workspace.
Outcome
This skill creates and maintains a repository-local memory root at .ai/memory/.
The memory is meant to be read before work starts and updated as durable facts
become clear.
Primary goals:
- Load high-value context before doing substantive work
- Persist stable repo knowledge discovered during the task
- Keep user-specific preferences separate from repo facts
More from jeff-tian/agent-skills
oidc-integration
Plan and implement OIDC and OAuth 2.0 integration for React or TypeScript frontends and Java or Spring Boot backends. Use whenever the user mentions OIDC, OpenID Connect, OAuth login, SSO, PKCE, authorization code flow, refresh tokens, JWT or JWKS validation, login callback pages, protected routes, Keycloak, Auth0, IdentityServer, Authing, multi-provider auth, or "add login" and "integrate IdP" style requests even if they do not explicitly say OIDC.
10markdown-output
>-
9tdd
Use when implementing any feature or bugfix, before writing implementation code
6cohosted-frontend-backend
Co-host a frontend SPA and backend API in the same server process or container — separate codebases in development, single entry point in production. Use when integrating React/Next.js/Umi frontends with Node.js (Fastify/Egg.js) or ASP.NET Core backends. Trigger on keywords like "co-hosted frontend backend", "single container deployment", "SPA + API same server", "add frontend to backend project", "integrate Next.js with .NET", "static export", "SSR takeover", etc.
4specflow-to-reqnroll
>-
3