菜单
Graph-based issue tracker for task coordination. Use when the project tracks work with beans — creating, querying, claiming, and closing tasks via CLI.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Cache and refresh remote git repositories under ~/.cache/checkouts/<host>/<org>/<repo> so future references can reuse a local copy. Use this skill when the user points you to a remote git repository as reference or you encountered a remote git repo through other means.
Test-driven development with red-green-refactor loop. Use when user wants to build features or fix bugs using TDD, mentions "red-green-refactor", wants integration tests, or asks for test-first development.
Python coding rules. Always use when working with Python, including editing, reviewing, or writing Python code, imports, code style, dependency injection, Pydantic, or Pytest.
Django coding rules. Always use when planning, writing, or reviewing changes in Django codebases.
Create new agent skills with proper structure, progressive disclosure, and bundled resources. Use when user wants to create, write, or build a new skill.
Build a throwaway prototype to flesh out a design before committing to it. Routes between two branches — a runnable terminal app for state/business-logic questions, or several radically different UI variations toggleable from one route. Use when the user wants to prototype, sanity-check a data model or state machine, mock up a UI, explore design options, or says "prototype this", "let me play with it", "try a few designs".
| name | beans |
| description | Graph-based issue tracker for task coordination. Use when the project tracks work with beans — creating, querying, claiming, and closing tasks via CLI. |
Graph-based issue tracker. Beans are nodes, dependencies are edges.
task-a3f2dd1c, epic-12345678, bug-deadbeef.beans types. Default: task, bug, epic.A blocks B means B can't start until A is closed.--parent.# Query
beans list # all beans
beans list --type bug --status open # filtered
beans ready # unblocked beans, sorted by priority
beans show <id> # full details with deps
beans search "query" # search title and body
beans stats # counts by status, type, assignee
beans graph # dependency tree
# Create and update
beans create "Title" # new task
beans create "Title" --type bug --body "Details" --parent <epic-id>
beans update <id> --title "New" --priority 0 --body "Updated"
beans close <id> --reason "Fixed in abc1234"
beans delete <id>
# Assignment
beans claim <id> --actor <name> # sets assignee + in_progress
beans release <id> --actor <name> # clears assignee + open
beans release --mine --actor <name> # release all claimed by actor
# Dependencies
beans dep add <blocker-id> <blocked-id> # blocker must close before blocked
beans dep remove <from> <to>
# Types
beans types # list configured types
beans types add spike --description "Time-boxed investigation"
beans types remove spike
# Structured output
beans --json list # JSON output on any command
beans --json --fields id,title,status list
beans schema # JSON schemas for all models
--json is a global option. Put it immediately after beans and before the subcommand. Order matters.
beans --json ready
beans --json search "output"
beans --json list | jq -r '.[] | [.id,.title,.status] | @tsv'
Do not put --json after the subcommand:
beans ready --json # wrong: No such option: --json
beans search "output" --json # wrong: No such option: --json
beans readybeans show <id>beans claim <id> --actor <name>beans close <id> --reason "Done in <commit>"beans ready before starting work.--json for structured output when integrating programmatically.