langsmith_run_evals
15025Skill Claude CodeCodex
Guidance for writing runevals.py scripts that use LangSmith evaluate() with a dataset ID.
not rated 5 7mo ago A 26 tokens
adk-workshop-builder
15026Skill Claude CodeCodex
Builds multi-language (Python, TypeScript, Go, Kotlin) autonomous agent and multi-agent system workshops using Google Agent Development Kit (ADK). Generates Coordinator-Worker agent architectures, tool bindings, Human-in-the-Loop (HITL) checkpoints, and trajectory evaluation labs.
not rated 5 29d ago A 63 tokens
original MIT
stet
15027Skill Claude CodeCodex
Use Stet to measure whether an AI coding change is safe to ship. Trigger on model comparisons, AGENTS.md or CLAUDE.md effectiveness, shared instruction, policy, skill, harness, tool-policy, reasoning, or runtime rollouts, repo eval setup, dataset building, regression detection, benchmarking, promote/rollback…
not rated 5 1mo ago A 116 tokens
decision-engine
15028Skill Claude CodeCodex
Pure decision-making logic. Maps (targetcfg, resourcestate, memory) → which variant, which counter, execute or skip, what to do with a drop, which trait/option to take, how the meta is trending. No I/O, no browser, no wallet, no LLM calls — just deterministic policy.
not rated 5 4mo ago A 2 tokens
report-generator
15029Skill Claude CodeCodex
Generate polished, enterprise dashboard-style reports as HTML files (never Markdown). Use this skill whenever the user asks for any kind of report or written technical deliverable — including status reports, Jira / sprint / release reports, implementation reports, investigation reports, bug reports, root cause…
not rated 5 2mo ago A 132 tokens
harness-engineering
15030Junhua-ele/codex-harness-skill
Skill Claude CodeCodex
Build or update a Harness Engineering knowledge base from a PRD, product idea, or existing project plan. Use when you needs to generate agent-ready project documentation, including PRD.md, DESIGN.md, FRONTEND.md, TECH-SPEC.md, ARCHITECTURE.md, docs/product-specs/, feature-list.json, AGENTS.md, README.md, or when a…
not rated 5 4mo ago A 97 tokens
telegram-admin
15031Skill Claude CodeCodex
Configure, inspect, and operate the local Codex Telegram bridge.
not rated 5 22d ago A 16 tokens
email-attention-flow
15032Skill Claude CodeCodex
Use for recall-sensitive scans of personal Gmail that must surface all emails the user personally owes a response to over a requested timeframe, across all relevant addresses, while still capturing other actionable mail, filtering bulk and list mail, honoring log.md, proposing standard-template replies from…
not rated 5 5mo ago A 79 tokens
original MIT
night-shift-setup
15033Skill Claude Code
This skill should be used when the user asks to "set up night shift", "add night shift workflow", "set up the agent loop", "initialize night shift", "add night shift to this project", or wants to configure autonomous overnight AI agent development using the Night Shift method by Jamon Holmgren.
not rated 5 6mo ago A 68 tokens
Skill Claude CodeCodex
Orchestrate a product idea into a complete, delivery-grade PRD workspace by running the PRD skills with stage gates and automatic feedback loops. Use when the user asks to run the full idea-to-PRD workflow, create a PRD from a rough idea, or coordinate multiple PRD drafting stages. Do not use for editing one isolated…
not rated 5 4mo ago A 88 tokens
original MIT
proprietary-leak-scan
15035Skill Claude CodeCodex
Use before committing to the public CTObot repo to catch company-specific or proprietary content that has drifted into the public files (AGENTS.md, SOUL.md, docs/, skills/, README). The agent reads the diff (or all files) and reasons about whether the content is generic and shareable or specific to our company…
not rated 5 1mo ago A 108 tokens
original MIT
report-compose
15036Skill Claude Code
A Paper Scout research-report workflow creates and archives reports in Lark, a collaboration and document platform. It covers document structure, diagrams, local media, delivery through Lark, and user notification.
not rated 5
changed 15d ago A 57 tokens
Skill Claude CodeCodex
Mantener sincronia entre CLI create-athenea-app y su template para scaffolding correcto.
not rated 5 2mo ago A 25 tokens
memory-management
15038Skill Claude CodeCodex
Two-tier memory system that makes agents true workplace collaborators. Decodes shorthand, acronyms, nicknames, and internal language so agents understand requests like a colleague would. AGENTS.md for working memory, memory/ directory for the full knowledge base.
not rated 5 7mo ago A 52 tokens
original Apache-2.0
reverse-engineer
15039Skill Claude CodeCodex
Reverse-engineer a website's internal APIs, encrypted endpoints, WebSocket streams, and obfuscated JavaScript. Activates when the target data isn't in the HTML, when the site uses encrypted CloudFront/CDN payloads, when real-time streaming data is needed, or when the scrape skill's Phase 0-1 finds API calls that are…
not rated 5 5mo ago A 96 tokens
original MIT
pipe
15040Skill Claude CodeCodex
Chain multiple AI agent skills into sequential pipelines. Output of each skill feeds as context into the next.
not rated 5 5mo ago A 22 tokens
gigaverse
15041Skill Claude CodeCodex
Enter the Gigaverse as an AI agent. Create a wallet, quest through dungeons, battle echoes, and earn rewards. The dungeon awaits.
not rated 5 7mo ago A 32 tokens
skill-audit
15042Skill Claude CodeCodex
Scans Claude Code skills for security risks before installation. Checks for prompt injection, credential harvesting, exfiltration patterns, hooks, excessive permissions, and settings modification. Supports GitHub URLs, local paths, and bulk scanning. Use when asked to audit, scan, or check a skill for malware or…
not rated 5 7mo ago A 67 tokens
original MIT
Skill Claude CodeCodex
Use this skill when investigating, adding, validating, or debugging external website profiles for the 99idea Playwright browser demo. It teaches how to probe selectors, classify failure modes, add config-driven profiles, and validate both heuristic and Gemini flows.
not rated 5 6mo ago A 54 tokens
original MIT
pdf2-md
15044Skill Claude CodeCodex
Convert academic PDF papers into clean, well-structured Markdown files using LLM vision. Use when the user provides a PDF paper and asks to extract text, tables, figures, or equations into Markdown, or mentions "parse paper", "PDF to markdown", "extract paper content", or similar.
not rated 5 6mo ago A 63 tokens
original MIT
doc-review
15045Skill Claude CodeCodex
A guide for creating password-protected pages where people can review documents and leave comments directly on them. Inline annotations are notes attached to specific parts of a document.
not rated 5 4mo ago D 65 tokens
kr
15046Skill Claude Code
Use when renaming Kotlin symbols (classes, methods, properties, parameters), moving Kotlin files to other packages, or analyzing Kotlin project structure (outline, find, deps, beans, hierarchy, unused). Triggers on rename, refactor, move, outline, find, analyze requests involving .kt files.
not rated 5 7mo ago B 63 tokens
original MIT
shortcuts-generator
15047Skill Claude CodeCodex
Generate macOS/iOS Shortcuts by creating .shortcut plist files. Use when asked to create, build, or generate Shortcuts, automate iOS/macOS workflows, or produce .shortcut files.
not rated 5 6mo ago C 44 tokens
original MIT
cmux-diff
15048Skill Claude CodeCodex
Start cmux-diff (VSCode-style changes panel) to view changed files and diffs in a sidebar layout. Use when user says "/cmux-diff", "show changes", "changes panel", or "diff viewer".
not rated 5 6mo ago A 52 tokens
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: