发现 Skills

Use when tracking work time and the user runs concurrent AI agent sessions, has overlapping time blocks, or works on multiple projects simultaneously in separate terminals. Also use when asked to produce daily work summaries, weekly capacity reports, or project-level time breakdowns where parallel sessions may inflate totals. Triggers: "time tracking", "daily summary", "hours per project", "weekly capacity", "parallel sessions", "concurrent agents", "overlapping blocks".

Drafts faculty application documents—research statements, teaching statements, DEI/diversity statements, and cover letters—tailored to a specific job posting, department, and institution. Use this skill whenever the user mentions faculty applications, academic job market, research statement, teaching philosophy/statement, diversity statement, cover letter for a professor position, or anything related to applying for tenure-track or teaching-focused academic positions. Also trigger when the user asks to tailor an existing application document to a new department or posting, or when reviewing/editing drafts of these documents.

React component patterns including functional components, hooks, state management, API integration, loading/error states, and event handlers. Use when creating or reviewing React components.

Git 2.49+ features including reftables, sparse-checkout, partial clone, git-backfill, and worktrees

Full MAF 1.3.0 migration reference guide, navigated by section. Load this Skill when you need API signatures, before/after code patterns, or detailed explanations of MAF 1.3.0 changes. The guide lives at guides/maf-1.3.0-migration-guide.md — this Skill is the smart index that tells you which section to read for a given question.

Scaffold Codex hooks hooks into a real project after auditing the repo, verifying the live official docs and schemas, inspecting the effective `codex_hooks` feature flag, and enabling it in project or user config if needed. Use when a user wants Codex hooks, `.codex/hooks.json`, managed Codex hook refreshes, repo-local lifecycle Bash hooks, or help with `SessionStart`, `PreToolUse`, `PostToolUse`, `UserPromptSubmit`, or `Stop`. Trigger on: Codex hooks, hooks.json, codex_hooks, .codex/config.toml, feature flag, managed hook scaffold. Do NOT use for Claude Code hooks, `.claude/settings.json`, Husky-only setups, or non-Codex agents.

Operate and apply GTFOBins — the curated list of Unix binaries exploitable for privilege escalation, file operations, shell spawning, and capability abuse. Use when performing Linux/Unix privilege escalation, post-exploitation enumeration, SUID/sudo/capabilities exploitation, or when the user needs to leverage a specific binary to escape restricted shells, read protected files, or gain a root shell. Covers all GTFOBins categories, key binary techniques, SUID enumeration, sudo -l parsing, capabilities exploitation, and integration with LinPEAS/LinEnum output.

Create Product Requirements Documents (PRDs) through structured interviews. Use when the user asks to create a PRD, write requirements, spec out a feature, or needs help defining what to build.

Use GitHub Spec Kit's /speckit.implement and /speckit.tasks to systematically build missing features from specifications. Leverages implementation plans in specs/, validates against acceptance criteria, and achieves 100% spec completion. This is Step 6 of 6 in the reverse engineering process.

WHAT: Automate HE wakeups, monitoring, until-green checks, and follow-through. Use when later thread continuation or goal-aware scheduling is needed.

Review, triage, and validate visual regression diffs. Use when the user wants snapshot-change analysis, layout regression evidence, Storybook diffs, Playwright screenshots, or Argos review.

Use when developing R packages that need to handle errors, warnings, and conditions properly. Covers: (1) cli_abort/cli_warn/cli_inform for throwing conditions with formatting, (2) Error call context with caller_env() and caller_arg(), (3) Input validation helpers that report the right function, (4) Error chaining with try_fetch() and parent argument, (5) Testing error conditions with testthat snapshots.

Runs cargo clippy with all-targets and pedantic lints on the touched files, groups findings by category, and applies cargo clippy --fix only after explicit per-file user approval while honoring existing

Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.

Force unrelated concepts together to discover emergent properties - "What if we treated X like Y?"

Delegate coding tasks to Claude Code (Anthropic's CLI agent). Use for building features, refactoring, PR reviews, and iterative coding. Requires the claude CLI installed.

Build complex AI systems with declarative programming, optimize prompts automatically, create modular RAG systems and agents with DSPy - Stanford NLP's framework for systematic LM programming

Convert large Markdown documents to Word .docx files using only the Python standard library, with proper Chinese government-document formatting (fonts, sizes, headings, tables).

Extract text from PDFs and scanned documents. Use web_extract for remote URLs, pymupdf for local text-based PDFs, marker-pdf for OCR/scanned docs. For DOCX use python-docx, for PPTX see the powerpoint skill.

Test Model Context Protocol (MCP) servers using the MCP Inspector CLI with correct syntax

Background knowledge for software development. Automatically provides context about architecture patterns (layered, microservices, clean architecture), design patterns (factory, observer, repository), API design (REST, GraphQL), and development tools when Claude is working on general programming tasks.

Skills for querying the KBase/BERDL Datalake via the MCP REST API. Use this when users want to explore KBase databases, list tables, get schemas, sample data, or run SQL queries against the KBase data lake. Triggers on mentions of KBase, BERDL, or requests to query biological/microbiome data stored in KBase.

Load automatically when planning, researching, or implementing ANY Medusa backend features (custom modules, API routes, workflows, data models, module links, business logic). REQUIRED for all Medusa backend work in ALL modes (planning, implementation, exploration). Contains architectural patterns, best practices, and critical rules that MCP servers don't provide.

Generate phonologically consistent constructed languages for fiction. Use when you need naming languages, alien speech, or fantasy tongues without deep linguistics knowledge.