发现优质的 AI Agent 技能

This skill provides Pi's structured approach to task management, quality assurance, and continuous self-improvement.

**适用于 Claude Projects / Custom Instructions**

**核心原则：用代码控制流程，不靠 AI 记忆。**

Mission: enforce OpenClaw's 2026-era security posture, block risky actions, and coach users toward safer workflows.

来自 openclaw/skills 技能

来自 openclaw/skills 技能

Interact with your local Logseq instance through its JavaScript Plugin API. This skill enables reading, writing, querying, and automating workflows in your Logseq graph.

Execute multi-step workflows with parallel tasks and multi-model orchestration via the Orchestrate platform.

- If the repo has `AGENTS.md` or `docs/agents/PROCESS.md`, read it for repo-specific rules.

- If the repo has `AGENTS.md` or `docs/agents/PROCESS.md`, read it for repo-specific rules.

Enforce a high-signal commit workflow and a human-written PR format. Keep global process rules as source of truth and make PRs reviewable by humans and agents.

Use Mobb MCP scan-and-fix behavior to identify security issues in a local repo and apply the generated patches. Follow the MCP workflow exactly, including file selection, pagination, and rescan rules.