Local guidance bank for coding agents - reusable rules, skills, and project guidance
# @morsa/guidance-bank `@morsa/guidance-bank` is a local tool for coding agents that stores persistent rules, skills, and reusable project guidance. It gives agents a stable guidance layer across sessions, projects, and tools. It gives you one durable place for reusable rules and skills across: - different agent providers - different projects - repeated sessions in the same project The goal is simple: - improve agent quality over time - reduce repeated prompting and repeated context reconstruction - save tokens by keeping stable guidance in a managed local guidance layer ## Quick Start Install globally: ```bash npm install -g @morsa/guidance-bank ``` Initialize once: ```bash gbank init ``` That is the whole manual setup. After that, your agent can work with the AI Guidance Bank during normal coding sessions. When a project has no bank yet, the agent can detect that and guide creation as part of the workflow. ## Why It Exists Agent guidance is usually fragmented. - Some rules live in `AGENTS.md`. - Some live in `.cursor`, `.cursor`, or `.codex`. - Some are project-specific. - Some should be shared across many repositories. - Most provider-native flows are still weak at generating a good long-lived bank from real project evidence. `@morsa/guidance-bank` solves that by giving the agent one canonical local AI Guidance Bank it can use across providers and across projects. It is designed for two kinds of guidance: - cross-agent reusable guidance shared between projects - project-specific guidance derived from the actual codebase and stack ## Supported Providers Current provider integrations: - Codex - Cursor - Claude Code ## What Happens Next After `gbank init`, the normal flow is intentionally lightweight: 1. You open a project in your agent. 2. The agent resolves AI Guidance Bank context for that project. 3. If a project bank does not exist yet, the agent can propose creating it. 4. The agent can then keep using, improving, syncing, and editing
Agent that generates comprehensive documentation, API references, architecture diagrams, and developer onboarding guides from existing code.
Agent configuration for systematic bug investigation that traces issues from error logs through the codebase to root cause with suggested fixes.
Agent for integrating third-party APIs including SDK setup, type generation, error handling, retry logic, and rate limit management.
Cursor's built-in autonomous coding agent that can make multi-file edits, run terminal commands, search the codebase, and iteratively build features with minimal human intervention.
Cloud-based autonomous coding agent that runs in the background on remote sandboxed environments, handling complex multi-step tasks while you continue working.
Cursor's multi-file editing agent within Composer mode that can create, edit, and delete files across your entire project in a single conversation.