harness-experiment
Run a harness change as a controlled experiment, with a small end-to-end eval, one change at a time, repeated runs, and looking at flipped tasks, request…
Find an Agent Skill for your task. Read the guide and copy the install command.
Updated 11-05 00:16
25,666 results
Run a harness change as a controlled experiment, with a small end-to-end eval, one change at a time, repeated runs, and looking at flipped tasks, request…
Audit how an agent harness runs long shell commands, covering foreground wait limits, moving to background without re-running, timeouts, completion…
Decide whether a piece of work should go to a subagent or stay in the main agent loop. Use before spawning subagents, when designing a multi-agent…
Audit how much context an agent's tools consume, through tool definitions and tool results, and decide whether to trim the catalog, add tool search, or add…
Build or upgrade complete Three.js browser games by coordinating gameplay, Mint MCP assets, graphics, UI, debugging, performance, QA, and release.
Build and upgrade Three.js visual systems for apps and games using Mint MCP assets, authored geometry, materials, lighting, shaders, VFX, technical art, and…
WHEN writing cross-agent messages, handovers or delegation briefs, or creating/editing agent skills; NOT for human-facing tutorials, marketing copy or code…
Package, publish, verify and update an agent skill through a Claude Code plugin marketplace, APM or the npx skills CLI. Use when preparing skill distribution…
Install and configure the Manifest LLM router plugin for OpenClaw. Use when the user says "/install-manifest", "install manifest", "add manifest plugin",…
Comprehensive tier assessment for an MCP SDK repository against SEP-1730. Produces tier classification (1/2/3) with evidence table, gap list, and remediation…
Research and draft a Specification Enhancement Proposal following the MCP SEP governance process
Search MCP PRs, issues, and discussions across the modelcontextprotocol GitHub org
Principal/Senior-level Tempo playbook for trace architecture, storage strategy, sampling-aware operations, tenant governance, and operating large-scale trace…
One unattended kanban foreman iteration: watch the backlog queue, dispatch the next operator-picked card to an isolated worker, collect completion evidence on…
Integrated audio generation skill. Supports ElevenLabs MCP-based TTS (32 languages), voice cloning (1-minute sample), multilingual dubbing (lip sync), and sound effect generation. Use for "voice generation", "TTS", "speech synthesis", "voice cloning", "dubbing", "narration", "sound effects", "AI voice" requests.
Midjourney v8.1 (2026.03 Alpha) dedicated image prompt builder. Collects context from a user's one-line natural language input plus AskUserQuestion presets and fine-tuning, and converts it into a keyword+`--파라미터` format based on the Midjourney official Parameter List.…
One-time meta skill that sets up a repository with all agent skills and walks through interactive setup. Installs internal skills from mohganji/skills,…
One-time setup skill that adds caveman as the default communication style to CLAUDE.md. Use when user wants caveman mode on by default or during bootstrap…
Defragment a work calendar through a tool-using agent — find the meeting debt, propose the consolidation, and (approval-gated) execute the moves. Use when…
Review what an LLM feature or agent actually puts in its context window — and find what's bloating, missing, or fighting itself. Use when asked to review a…
Choose the right LLM for a task by trading off quality, cost, latency, and constraints. Use when asked which model to use, whether to upgrade/downgrade a…
Review an existing Retrieval-Augmented Generation system and find why it underperforms. Use when asked to review or audit a RAG pipeline, diagnose…
Write a standout rental application / cover letter to a landlord or letting agent. Use when asked to write a rental application, a letter to a landlord, a…
Translate and sync bilingual user documentation between docs/zh/ and docs/en/ following the source-of-truth rules in docs/AGENTS.md.