Skill Codex
This skill should be used when the user asks about libraries, frameworks, API references, or needs code examples. Activates for setup questions, code generation involving libraries, or mentions of specific frameworks like React, Vue, Next.js, Prisma, Supabase, etc.
Skill Codex
Use gh (GitHub CLI) to fetch and process GitHub issues, pull requests, repositories, and other GitHub resources. Use when the user provides a GitHub URL (https://github.com/), asks about GitHub issues/PRs, or mentions gh commands.
Skill Claude CodeCodex
Orchestrate PyTorch test file refactoring to decouple tests from specific hardware accelerators. Use this skill whenever the user asks to refactor, decouple, or reorganize a PyTorch test file for cross-accelerator compatibility, or when they ask to apply the test decoupling workflow to a specific test file. Triggers…
Skill Claude CodeCodex
Use when the pytorch-test-refactoring orchestrator emits "schedulecron" or "needagent" with phase "cidebug", when a CronCreate cron job fires and you need to run orchestrator.py --ci-check, or when the user says "look after the ci", "monitor ci", "check ci status", "watch the pr", or "debug ci failures".
Skill Claude CodeCodex
Refactor PyTorch test files to decouple tests from specific hardware accelerators using three strategies: CPU-only (standalone classes with instantiateparametrizedtests), device-agnostic (device-parametrized classes with instantiatedevicetypetests), and device-specific (single-accelerator standalone classes). Pay…
Skill Claude CodeCodex
Review PyTorch test refactoring for correctness and completeness against the decoupling standards defined in the refactor-test-decoupling skill. Accepts a test file path (whole-file review), a PR URL, a git diff, or a branch name. Use this when asked to review a test refactoring PR, check a test decoupling change…
Skill Codex
Use when asked to find open, unassigned, or claimable vllm-project/vllm issues to work on, or when judging whether a vLLM issue can be reproduced or verified on this machine's hardware (GPU model, compute capability, accelerator vendor, GPU count).
Skill Codex
Use when writing, fixing, or reviewing a Mermaid diagram in markdown or docs — flowcharts, sequence diagrams, state machines, class/ER diagrams, gantt, mindmap, timeline and friends — including "mermaid syntax error", "diagram won't render", "this diagram is unreadable", or any request to draw a process, architecture…
Skill Codex
Use when a task needs live web information — current facts, docs, releases, pricing, news — or the full text of a URL as clean markdown, especially when webfetch is blocked, fails to resolve the hostname, or returns raw HTML, or when the session has no built-in web search.
Skill Claude CodeCodex
A skill for teaching technical topics and explaining how or why code and systems work.
Skill Codex
A repository research tool for understanding the internals of public GitHub projects. It helps identify the files, classes, and modules responsible for a mechanism in an unfamiliar codebase.
Skill Codex
Use when asked to find open, claimable pytorch/pytorch issues to work on for a module configured in this skill's target file (inductor, fsdp2, and the like), or when judging whether a PyTorch issue is still present, already claimed, or reproducible and verifiable on this machine.
Skill Claude CodeCodex
Turns the usage data this repo's collectors record into a dated insight — which skills each prompt triggered, tokens per skill run, whether declared sub-skills fired, which scripts fail and which files nobody reads — and records a keep / update / retire decision per skill. Covers Claude Code (hooks) and dsh (session…
Hook Claude Code
One of 11 in hooks.json
Runs when a session starts, running an inline shell check. From KarhouTam/agent-skills.
Hook Claude Code
One of 11 in hooks.json
Runs when you submit a prompt, before the agent sees it, running an inline shell check. From KarhouTam/agent-skills.
Hook Claude Code
One of 11 in hooks.json
Runs to expand your prompt before the agent sees it, running an inline shell check. From KarhouTam/agent-skills.
Hook Claude Code
One of 11 in hooks.json
Runs before the agent uses a tool for Skill tool calls, running an inline shell check. From KarhouTam/agent-skills.
Hook Claude Code
One of 11 in hooks.json
Runs after a tool call finishes for Skill, Read, Edit, Write, MultiEdit, NotebookEdit, Bash, Agent and Task tool calls, running an inline shell check. From KarhouTam/agent-skills.
Hook Claude Code
One of 11 in hooks.json
Runs when a tool call fails for Skill, Read, Edit, Write, MultiEdit, NotebookEdit, Bash, Agent and Task tool calls, running an inline shell check. From KarhouTam/agent-skills.
Hook Claude Code
One of 11 in hooks.json
Runs when a subagent starts, running an inline shell check. From KarhouTam/agent-skills.
Hook Claude Code
One of 11 in hooks.json
Runs when a subagent finishes, running an inline shell check. From KarhouTam/agent-skills.
Hook Claude Code
One of 11 in hooks.json
Runs when the agent finishes a response, running an inline shell check. From KarhouTam/agent-skills.
Hook Claude Code
One of 11 in hooks.json
Runs on the StopFailure event, running an inline shell check. From KarhouTam/agent-skills.
Hook Claude Code
One of 11 in hooks.json
Runs when a session ends, running an inline shell check. From KarhouTam/agent-skills.
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: