菜单
Apply and teardown declarative YAML templates for agents and workflows. Use for deploying agent stacks from .agentd/ directories, validating templates, and cleaning up resources.
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。
Store, search, and manage shared knowledge through the memory service. Use for persisting information across agent sessions, semantic search over stored knowledge, and inter-agent knowledge sharing.
Browse, read, and author per-project knowledgebase documents through the knowledge service CLI. Use for creating, updating, listing, and reading markdown documents stored against a project, and reconciling DB rows against disk.
Cut a new agentd release. Suggests the next semver version from conventional commits (or accepts an explicit `x.y.z`), bumps the workspace version, regenerates the changelog with git-cliff, then commits, tags, and pushes the `v*` tag that triggers the Release workflow. Use when the user runs /release or asks to publish/cut/tag a new version.
Create a new GitHub plan issue with the "Plan: <title>" format, appropriate labels, and structured body. Use when the user wants to capture a high-level plan for future work.
Ask the human user a question during workflow execution and check for answers. Use for agent-to-human Q&A, collecting decisions, and reacting to human responses via ask_response workflow triggers.
Branch stacking and PR management with git-spice for the agentd project. Use when creating stacked branches, submitting PRs, syncing with upstream, resolving rebase conflicts, or navigating a branch stack.
| name | agent-apply |
| description | Apply and teardown declarative YAML templates for agents and workflows. Use for deploying agent stacks from .agentd/ directories, validating templates, and cleaning up resources. |
Skill for deploying and tearing down agentd resources from declarative YAML templates.
# Apply an entire .agentd/ directory (agents first, then workflows)
agent apply .agentd/
# Apply a single template file
agent apply .agentd/agents/worker.yml
agent apply .agentd/workflows/issue-worker.yml
# Dry run — validate without creating anything
agent apply .agentd/ --dry-run
# Custom timeout for waiting on agents to become Running
agent apply .agentd/ --wait-timeout 120
Running status (default 60s timeout)# Teardown entire directory (workflows first, then agents)
agent teardown .agentd/
# Teardown a single resource
agent teardown .agentd/agents/worker.yml
# Dry run — show what would be deleted
agent teardown .agentd/ --dry-run
.agentd/
├── agents/
│ ├── worker.yml
│ ├── reviewer.yml
│ └── planner.yml
└── workflows/
├── issue-worker.yml
└── pr-reviewer.yml
# .agentd/agents/worker.yml
name: worker
working_dir: "."
shell: /bin/zsh
worktree: false
model: claude-sonnet-4-6
system_prompt: |
You are a worker agent for the project.
Focus on implementing features and fixing bugs.
env:
RUST_LOG: debug
GITHUB_TOKEN: "${GITHUB_TOKEN}"
Required: name, working_dir, model
Optional: shell, worktree, system_prompt, env
# .agentd/workflows/issue-worker.yml
name: issue-worker
agent: worker
source:
type: github_issues
owner: geoffjay
repo: agentd
labels:
- agent
state: open
poll_interval: 60
enabled: true
prompt_template: |
Work on issue #{{source_id}}: {{title}}
URL: {{url}}
Labels: {{labels}}
{{body}}
Required: name, agent, source, prompt_template
Optional: poll_interval (default 60), enabled (default true), tool_policy
{{source_id}} — GitHub issue/PR number{{title}} — issue/PR title{{body}} — issue/PR body text{{url}} — GitHub URL{{labels}} — comma-separated label list