В каталоге 185 367 навыков в 14 категориях.
Implement OpenAI Harness Engineering practices in any repository. Use when setting up or refactoring agent-first workflo...
Structured SE task evaluation using 15 benchmark definitions from claude-code-harness research
Evaluate a repo agent harness (AGENTS.md, rules, skills, skill refs) for broken paths/commands, redundant instructions,...
Use when evaluating a reference-harness capability for adoption, rejection, deferral, or synchronization. Inspection alo...
Improve agent reliability over time — diagnose why agents fail and fix the setup. Triggers on: agent keeps failing, same...
Run a harness change as a controlled experiment, with a small end-to-end eval, one change at a time, repeated runs, and...
Expert-level Harness template types, runtime inputs, expression language, pipeline patterns (CI/CD, GitOps, Canary, Blue...
Use when mapping PonyBunny's current state against harness-first best-practice gaps and turning those gaps into actionab...
Seed minimal agent harness v0 — manifest, eval checks stub, governance. AUTO-INVOKED after project-setup or retroactive-...
HR hiring. Searches .harness/shared/HR-Resource candidates, installs selected worker skills into .claude and .codex, and...
Use when the coding agent repeatedly misses the same class of bug, verification, context, security, UX, session-continui...
Use when turning repeated failures, weak evals, weak handoffs, or trace blind spots into concrete improvements to the de...
Project setup with environment check, file generation, SSOT sync, and validation. Use when user mentions '/harness-init'...
Keycloak OIDC integration with Harness pipelines, EKS IRSA, service account authentication, and realm-as-code patterns
Run a GEPA learning cycle via `metaharness learn` (upstream ADR-235, metaharness@0.3.0) — optimizes a harness genome aga...
Measures what this `.claude/` costs in context — every session and per skill, agent, rule and hook — through `ArchHook.j...
Long-running task loop using /loop (Claude Code dynamic mode) and ScheduleWakeup to re-enter with fresh context on each...
Harness MCP (Model Context Protocol) server integration for AI-powered CD operations, pipeline management, Git repositor...
Scaffold a custom AI agent harness via `metaharness new <name> --template <id> --host <id>`. Defaults to DRY-RUN (no wri...
Analyze the protocol layer between agent harness and LLM model. Use when (1) understanding message wire formats and API...
OPS environment monitor. Tracks build commands, mapped service servers, ports, logs, and emergency events.
OPS-only dev-mode visual API binding for React/Next.js or static HTML UI elements and REST endpoints.
Plans and tracks tasks in .claude/harness/plans.json. Use when creating plans, adding tasks, updating markers, checking...
HAR: Research-backed, team-validated task planning, Plans.md management, progress sync. Trigger: create a plan, add task...
Harness Platform administration including delegates, RBAC, connectors, secrets, templates, policy as code (OPA), user ma...
Publish this repo's harness into the mvp-plugin (the mvp-harness submodule) — copy the shippable skills and agents, run...
Executa um fluxo direto de 5 passos (escopo, inspeção, classificação, pontuação/notas, entrega) para avaliar o harness d...
Invoke an already-built Built Harness on one or more queries, without building a new one. Use when the operator points a...
Runs iterate-until-pass loops with ralph-worker agents in a worktree. Use when running [ralph]-marked tasks or looping u...
Turn a repeated agent or prod failure into a permanent harness improvement (test/hook/script + optional AGENTS line + RU...
Parse harness output (Surefire/Failsafe XML, JaCoCo XML, PIT XML, Checkstyle/SpotBugs XML, OpenAPI diff JSON, dependency...
Resource manager. Maintains hired-worker wording, aliases, keyword mapping, and hire recommendations.
コード・プラン・スコープを多角的にレビュー。品質の番人、参上。Use when user mentions reviews, code review, plan review, scope analysis, security, perfo...
对项目的 AI 编码工作流(Harness)做轻量体检:从任务理解、受控执行、变更验证、可靠交付、学习沉淀五个维度评估机制是否存在与真正被使用,产出带证据状态标注的报告(纯 Markdown,零运行时依赖)。触发:"体检一下工作流"、"评估...
Multi-angle code and plan review with security, scope, and UI profiles. Use when reviewing code, plans, PRs, or running...
Use when choosing workflow/tooling across Pi, Claude Code, Supacode, model roles, delegation, subagents, background agen...
Run WaveCap-SDR test harness with automated parameter sweeps and validation. Use when regression testing, validating aud...
Scan a reference harness for material upstream drift and missing capabilities.
Use when running Plans.md tasks on a scheduled cadence (fresh context per wake-up, sprint-contract flow, plateau detecti...
5-dimension harness readiness scorecard from `metaharness score <path>`. Returns harnessFit / compileConfidence / taskCo...
Use for Crewrift-specific harness sdk protocol recipes when optimizing a player.
[Quality] Use when a workflow step or the user asks for an agent quality harness: feedforward guides and feedback sensor...
Initializes and configures Harness in a project — CI, memory, duo workflow, Codex. Use when setting up a new project or...
HAR: Project init, tool setup, agent config, memory setup, skill mirror sync. Trigger: setup, init, new project, CI/Code...
Use for a substantive comparison of skill triggers, behavior or curation choices. Scale narrow overlap questions to the...
Quick upstream-drift status across all reference_harnesses submodules — what changed since our pinned commits. Fetches u...
Creates or updates pipeline step schemas in the harness-schema repository. Use when the user wants to add a new step, mo...
Harness Engineering 第一阶段:扫描现有项目,生成 AGENTS.md(目录文件)和完整的 docs/ 知识库结构。 当用户想要"为项目添加 agent 支持"、"让 AI 更好地理解我的项目"、"开始 harness...
Harness Engineering 第一阶段第二步:深度分析项目代码,填充业务解决方案、架构、约定、技术决策和质量标准等 docs/ 知识库内容。 在 harness-step1-create-agents-md 创建好目录骨架之后使...
Harness Engineering 第二阶段:建立跨 session 状态管理,解决 agent 每次对话失忆的问题。 创建 tasks.json(任务清单)、progress.md(进度记录)、init.sh(环境初始化脚本)三个文件...
Take stock of the whole harness in one report and one approval, then apply on a branch (ハーネスの棚卸し).
Use when user invokes /harness-sync or wants to sync AI tool configurations across Claude Code, Cursor, and GitHub Copil...
HAR: Sync Plans.md with implementation. Drift detect, marker update, retrospective. Trigger: sync-status, where am I, ch...
Testing orchestrator for autonomous harness. Runs comprehensive tests (unit, integration, E2E), coordinates with validat...
Harnessダッシュボードを表示。現状を一目で把握できる司令塔。Use when user mentions '/harness-ui', dashboard, monitoring, or UI display. Do NOT load...
Harnessを安全にアップデート。バックアップ付きで安心安全。Use when user mentions '/harness-update', update harness, update version, or template up...
Use when user invokes /harness-validate or wants to check whether a harness.yaml file is valid according to the Harness...
Executes plans.json tasks — solo, parallel, or breezing team mode. Use when implementing tasks or running the work loop.
HAR: Execute Plans.md tasks from single task to full parallel team run. Trigger: implement, execute, do everything, bree...
Make a Claude-harness worktree (.claude/worktrees/*) actually runnable - junction node_modules, then obey the one-checko...
Designs and improves fuzzing harnesses for C/C++ and Rust. Covers mapping raw bytes onto a target API, generating struct...
Techniques for writing effective fuzzing harnesses across languages. Use when creating new fuzz targets or improving exi...
Use when the user wants to verify that the evolved agent's score is stable and reliable. Runs evaluation multiple times...
Use when the user is done evolving and wants to finalize, clean up, tag the result, or push the optimized agent.
Use when the user wants to run the optimization loop, improve agent performance, evolve the agent, or iterate on quality...
Use when the user wants to check dataset quality, diagnose eval issues, or before running evolve. Checks size, difficult...
Use when the user wants to set up the evolver in their project, optimize an LLM agent, improve agent performance, or men...
Use when the user asks about evolution progress, current scores, best version, how many iterations ran, or whether the l...
Operate google-research/harness2 (Harness²), the Apache-2.0 framework for recursive agent harnessing: refine the prompts...
Harness²: Recursive Agent Harnessing (Google Research / arXiv:2610.14224). Dynamically adapts and self-improves the agen...
Generate project-adapted enterprise-configuration assets (CI, scanning, supply-chain, content) into /harnesses, record t...
Create a haro response with structured process, quality checks, and system integration
Yapılandırılmış süreç, kalite kontrolleri ve sistem entegrasyonu ile bir haro response oluşturun
MoneyPrinterTurbo — AI video generator từ topic/keyword: tự viết script, lấy stock footage, tạo subtitle, thêm nhạc nền....
Evidence-backed critic for code, plans, and analysis using structured gap analysis, multi-perspective review, calibrated...
AI-native E2E testing orchestration for Claude Code. Use when creating, running, or managing end-to-end test scenarios w...
Harting industrial connector MPN encoding patterns, series decoding, and handler guidance. Use when working with Harting...
Promote .agents knowledge.
Harvest integration for AI agents via ClawLink — browser login, no API key setup. Connect Harvest through ClawLink's hos...
Run an adaptive harvest — directed elicitation that extracts, deepens, and synthesizes what people think about a topic....
Capture durable session learnings, stage for human promotion to 05-knowledge/lizard, and propose skill/CLAUDE.md patches...
Report where a session missed moo's standard — the agent asks only while stating intent and shape, then works on its own...
Agent-callable Harvest tools — track time (log hours, start/stop timers), and manage the projects, clients, contacts, an...
コードベース全体を棚卸しし、ドキュメント化されていない暗黙知を洗い出して docs/ / .claude/rules/ / .claude/knowledge/ に直接反映する。 初期導入時や定期棚卸し用のオンデマンドスキル。 「/harv...
Multi-page deep crawling - documentation sites, wikis, knowledge bases
Web change monitoring - track changes on pages, detect updates, changelog diffs
Plan harvest operations with timing, labor, and storage. TRIGGERS - Use when user needs help with harvest-plan related t...
Single page smart extraction - articles, docs, blog posts to clean markdown
Structured data extraction - tables, pricing, products, API endpoints with schema
Pull reusable lessons from agent/PROMOTE.md across all projects registered in PROJECTS.md and propose them as additions...
Use HasData APIs for web scraping and structured web data extraction.
Command-line access to search, scraping, and structured web data.
Hash kırma — MD5/SHA/bcrypt format tespiti, hashcat/john kullanımı, wordlist stratejisi
Use when you need to recover plaintexts from password hashes on an authorized engagement, such as cracking NTLM hashes f...
Use when a user asks to check, identify, or characterize a file hash (MD5, SHA-1, SHA-256). Chains VirusTotal and OTX, o...
Compute the SHA-256 hash of a string or file
Building LLM-powered React applications with the Hashbrown library. Use when the user asks to (1) Build generative UI wh...
Crack hashes offline with hashcat: choose the right -m mode, run dictionary, rule, mask and hybrid attacks, and manage t...
Design CLI tools in the style of Mitchell Hashimoto, founder of HashiCorp. Emphasizes consistent command patterns, helpf...
Computes cryptographic and fuzzy fingerprints for malware samples: MD5/SHA-1/SHA-256 for exact identity, plus import has...