From hone
Generates AGENTS.md project documentation with feedback instructions, architecture, conventions, and discovered commands. Analyzes package.json, configs, directory structure, CI workflows. Use for new projects or doc updates.
How this skill is triggered — by the user, by Claude, or both
Slash command
/hone:agents-mdThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Generate AGENTS.md documentation for this project:
Generate AGENTS.md documentation for this project:
Check $ARGUMENTS for overwrite intent (look for --overwrite, overwrite, force, regenerate, etc.).
If AGENTS.md already exists and overwrite was NOT requested:
Analyze the project by reading:
package.json (dependencies, scripts, name, description)tsconfig.json, jest.config.*, vitest.config.*, .eslintrc*, tailwind.config.* etc.src/, lib/, test/, components/, .github/workflows/)README.md for project description.agents/ directory files if any.github/workflows/*.yml for CI/CD infoDockerfile, docker-compose.yml if presentGenerate AGENTS.md with this structure:
<!-- BEGIN GENERATED: AGENTS-MD -->
# AGENTS.md
Learnings and patterns for future agents working on this project.
## Feedback Instructions
TEST COMMANDS: [<discovered test commands>]
BUILD COMMANDS: [<discovered build commands>]
LINT COMMANDS: [<discovered lint commands>]
FORMAT COMMANDS: [<discovered format commands>]
## Project Overview
PRIMARY LANGUAGES: [<languages>]
<brief description>
## Build System
BUILD SYSTEMS: [<build tools>]
<details>
## Testing Framework
TESTING FRAMEWORKS: [<test runners>]
<details>
## Architecture
ARCHITECTURE PATTERN: <pattern description>
<details>
## Deployment
DEPLOYMENT STRATEGY: <strategy>
<details>
---
_This AGENTS.md was generated using agent-based project discovery._
<!-- END GENERATED: AGENTS-MD -->
Discover feedback commands from:
package.json scripts (test, build, lint, format, check, typecheck)run: steps)Create .agents/ directory if it doesn't exist, and generate detail files:
.agents/languages.md - Language stack details.agents/build.md - Build system details.agents/testing.md - Testing framework details.agents/architecture.md - Architecture patterns.agents/deployment.md - Deployment strategyIf AGENTS.md exists and overwrite was requested:
<!-- BEGIN GENERATED --> / <!-- END GENERATED --> markersWrite AGENTS.md to project root.
npx claudepluginhub oskarhane/hone-ai --plugin honeCreates and maintains concise AGENTS.md / CLAUDE.md files for projects. Analyzes toolchain, linters, and conventions to produce minimal agent-facing documentation under 100 lines.
Generates and maintains AGENTS.md files for AI agents using shell scripts that detect project structure, extract commands and CI rules, and validate content freshness.
Analyzes repository structure to generate or update AGENTS.md contributor guides for AI agents. Supports single-repo and monorepo layouts with update mode that preserves custom sections.