177 skills · sorted by relevance
Apply consistent rules when creating or editing Claude Code skills. It covers concise instructions, usage conditions, trigger registration, and tool permissions, with additional guidance for skills that call scripts for MCP operations.
Create Claude Code skills from reusable scripts or workflows that connect external tools. It guides the structure of SKILL.md, supporting files, execution instructions, and optional rules that suggest when a skill should be used.
Use Shapely to create points, lines, and polygons from coordinates or text descriptions of shapes. Calculate areas and distances, test spatial relationships, combine or transform shapes, and check or repair invalid geometry, with results returned as JSON.
Audit sensitive code and dependencies for security weaknesses, then verify fixes with tests and repeat checks. Reports identify severity, file locations, and remediation steps for issues involving authentication, data protection, injection attacks, and configuration.
Guides code searches by choosing tools for meaning, syntax structure, or literal text. Uses LEANN for conceptual questions, AST-grep for code patterns, Morph for text searches, and built-in Grep for simple queries or when Morph credentials are unavailable.
Choose a code search tool based on whether you need symbols, structural patterns, conceptual answers, or exact text. The rules guide agents toward relevant files before reading full contents, helping limit unnecessary search output.
Choose code search tools based on the question: AST-grep for code structure, LEANN for meaning, and Grep for exact names. Read complete files after locating relevant matches to limit unnecessary output and token use.
Organize agent skills around domain routers that map requests to specific commands. It guides router selection, automatic activation alongside related skills, and the separation between broad activation rules and precise command matching.
Review code changes from several perspectives using parallel specialist agents. It combines findings on code quality, design alignment, and change risks into a final verdict, prioritized actions, and a distinction between blocking and non-blocking issues.
Resume work from a handoff document by reviewing linked materials and checking the current code against recorded changes. The skill proposes next steps, creates a prioritized task list, and assigns approved work to specialist agents.
Investigate how an existing codebase works and document its components, connections, and file references. Combine current code findings with historical notes to produce a research document grounded in specific paths and line numbers.
Research library usage, recommended approaches, and general topics using external documentation and web sources. Choose a brief or thorough investigation, then organize examples, pitfalls, comparisons, and source links into a research document or implementation handoff.
Researches library documentation, implementation practices, and specific web pages using Nia, Perplexity, and Firecrawl through MCP tools. Produces a handoff document with findings, available code examples, recommendations, pitfalls, and source links for another agent.
Explore code with RepoPrompt CLI using directory trees, function signatures, targeted searches, and selected line ranges. Assemble relevant context for AI tasks, automatically select related files, and save repeatable command workflows while limiting how much code is read.
Investigate an unfamiliar codebase and create a handoff document covering its structure, technologies, implementation patterns, and contribution rules. Findings include source file references, issue and pull request templates, testing requirements, and practical points to check before contributing.
Save working solutions, design decisions, code patterns, and failed approaches for future recall. The skill classifies entries, extracts tags, stores them in PostgreSQL with numerical text representations, and returns a record ID.
Prepare software releases through security checks, end-to-end tests that cover the full system, and final review. After approval, coordinate version updates, a changelog, and release notes, with blocking issues stopping the workflow.
Use btca ask to inspect reference SDKs, libraries for connecting to services, when building features or debugging provider-specific issues. Compare streaming, tool calling, structured output, and validation approaches in Vercel AI SDK, Anthropic SDK, and Zod.
Refactor existing code through analysis, planning, implementation, review, and testing. Use small, verifiable changes to extract modules, reorganize code, or improve architecture while preserving existing behavior and checking for regressions.
Search stored learnings from past sessions by meaning or keywords. Retrieve relevant entries in full, along with their type, confidence, and session context, to revisit earlier findings about development patterns, installation, or errors.
Search previous work records to recover useful decisions, successful approaches, and failed attempts. Find context in handoffs, plans, and build or test histories, then use those findings when tackling similar tasks or recurring errors.
Integrate QLTY code checks into development after new files, substantial edits, and before commits. It provides commands for checking selected files, applying automatic fixes, and formatting code, with guidance on when to run them.
Check code quality with qlty CLI, a command-line tool for multiple languages. Review changed or selected files, automatically fix supported issues, format code, and measure complexity and duplication to identify areas that need attention.
Turn mathematical claims into Lean4 proofs, checked by a theorem prover. A five-phase workflow researches existing results, designs the proof, tests candidate lemmas, implements it, and reports remaining proof gaps and any custom axioms.