Loading...
Loading...
Found 1,462 Skills
Self-tuning question sensitivity + developer psychographic for gstack (v1: observational). (gstack)
All network operations must be processed through this skill, including: searching, web scraping, post-login operations, network interactions, etc. Trigger scenarios: When users request to search for information, view web page content, access websites requiring login, operate web interfaces, scrape social media content (Xiaohongshu, Weibo, Twitter/X, etc.), read dynamically rendered pages, and any network tasks that require a real browser environment.
Orchestrate a multi-ticket feature: one /implement agent per sub-issue; orchestrator sequences, relays, merges.
STOP — if you were dispatched as a subagent to execute a briefed task, this skill is not for you: it governs the top-level session only. Do the task you were briefed with instead. Use when you ARE that top-level session, driving a multi-step task through a fleet of specialized subagents — settling how the task will run, decomposing the work, and dispatching explorers, investigators, implementors, fixers, and reviewers rather than editing many files or running long gate loops yourself. Use ONCE at the start of a non-trivial task, before the first dispatch, whatever the deliverable is: a code change, a plan, a plan review, or a code review. Never reload it: once loaded it governs the whole task; re-opening the gate is a decision, not a reload. Symptoms it prevents: committing to a supervision level or round cap the operator never chose, the orchestrator becoming the implementor, serial dispatch of independent work, parallel writers colliding, vague briefs, and relaying raw subagent dumps.
Compact the current conversation into a single, detailed handoff message — everything that happened, why it happened, and what's left — output in a code block so it can be copy-pasted into a fresh agent session. Use when hitting context limits, switching focus, ending a work session, or partitioning a task across fresh contexts.
Use when working on complex multi-step tasks, when a session is getting long (40+ tool calls), when the agent starts ignoring rules it followed earlier, when conventions drift, when output quality seems to degrade, or after any context compaction event. Prevents long-session corruption AND context compaction amnesia through behavioral self-enforcement.
Use when adding Bale (بله) support to Hermes Gateway. Connect AI agent to Persian messenger via free Bot API.
In chat-only mode, reply like a chat assistant without changing the current code base.
Use only when explicitly invoked for a concrete output or authorized file or code change where latency is worth reduced exploration, accuracy, completeness, and proof breadth.
Use when observing and rechecking an ongoing process or state until a requested condition, deadline, or meaningful change occurs without mutation.
Set up OpenClaw locally and run it reliably on a Mac mini for private, always-on local agent workflows.
Sharpen the prompt before you send it, whether the one-shot instruction you're about to hand a coding agent or the system prompt your application ships. Use when the user says "optimize this prompt", "what's wrong with this prompt", "write or rewrite the system prompt my app ships", or "/promptkit".