agent-skills

Convert URLs or local files (PDF/DOCX/PPTX/HTML) to Markdown, with optional summarization. Single or batch processing. Uses markitdown for intelligent extraction, falls back to curl+pandoc. Use when fetching web pages as markdown, converting documents for analysis, or producing quick summaries.

Use `uv` instead of pip/python/venv. Run scripts with `uv run script.py`, add deps with `uv add`, use inline script metadata for standalone scripts.

Adversarial code reviewer. Use when you need a quality agent to critically review code, a diff, test output, or a pull request — looking for bugs, edge cases, correctness failures, lazy shortcuts, and pattern violations. Works standalone (against any code/diff/worktree) or with GitHub PR mechanics (inline comments, merge decision) when a PR number is provided. Triggers: "bart", "review this", "adversarial review", "code review", "review PR", "find bugs", "QA", "quality check".

Use this skill to orchestrate multiple parallel sub-agents (Ralph/Bart) across isolated git worktrees to burn down a TODO.md backlog. Trigger this when asked to implement multiple independent features simultaneously, dispatch a "wave of agents", or perform "multi-agent orchestration". This handles complex dependency analysis, model spread, and feedback accumulation, which standard dispatch skills cannot.

Launch Claude Code CLI as a headless sub-agent in a tmux pane, monitor its stream-json JSONL output, and poll for completion. Use when delegating a well-defined coding task (bug fix, feature, refactor) to a Claude agent subprocess with live observability. Triggers include "run claude agent", "delegate to claude", "claude sub-agent", or any request to run claude headlessly and monitor it.

Launch GitHub Copilot CLI as a headless sub-agent in a tmux pane, monitor its JSONL output stream, and poll for completion. Use when delegating a well-defined coding task (bug fix, feature, refactor) to a Copilot agent subprocess with live observability. Triggers include "run copilot agent", "delegate to copilot", "copilot sub-agent", or any request to run copilot headlessly and monitor it.

Launch Gemini CLI as a headless sub-agent in a tmux pane, monitor its stream-json JSONL output, and poll for completion. Use when delegating a well-defined task to a Gemini agent subprocess with live observability. Triggers include "run gemini agent", "delegate to gemini", "gemini sub-agent", or any request to run gemini headlessly and monitor it.

Use this skill when you need to act as the Planning Agent (Lisa). This persona focuses on logic, structure, architecture, and task breakdown.

Use this skill when you need to act as the Release Agent (Lovejoy). This persona focuses on ceremony, shipping, learning, and communication.

Use this skill when you need to act as the Product Agent (Marge). This persona focuses on empathy, user needs, and product definition.

Trigger native web search. Use when you need quick internet research with concise summaries and full source URLs.

Launch a pi coding agent as a sub-process in a tmux pane, monitor its progress via JSONL streaming, and report results. Use when you need to delegate a well-defined task (bug fix, feature, refactor) to a separate agent instance with live observability. Triggers include "launch agent", "delegate to sub-agent", "run this in another agent", "Ralph loop", "TDD in a sub-agent", or any request to run pi in a tmux pane and monitor it.

Build Agent — Red-Green-Refactor TDD executor that burns down a TODO.md task list. Reads TODO.md as the plan, implements each task using strict red-green TDD (write failing test → minimal code to pass → refactor), updates task status in TODO.md on completion, and creates an Atomic Commit Protocol (ACP) commit per task. Automatically progresses to the next task and repeats until the TODO.md is fully burned down. Uses INVEST task decomposition, Farley per-test quality checklist, and Conventional Commits.

Guide on applying Gojko Adzic's Specification by Example (BDD) principles. Use this skill to write new feature files, refine acceptance criteria, or evaluate existing Gherkin scenarios using the Adzic Index. Triggers on 'adzic', 'bdd', 'specification by example', 'gherkin', 'feature file'.

Guide on applying Dave Farley's Properties of Good Tests (TDD). Use this skill to write robust unit tests, practice strict Red-Green-Refactor, or evaluate a test suite using the Farley Index. Triggers on 'farley', 'tdd', 'test driven development', 'unit tests', 'test quality'.

Review tests for design quality using test design guidelines.

Discover, audit, update, and organise skills across all agent skill directories. Use when asked to list installed skills, find duplicates or broken symlinks, update a skill with session learnings, promote a project skill to the personal collection, or audit skill health. Triggers on "list skills", "update skill", "skill audit", "promote skill", "skill inventory", "what skills do I have", "sync skills", or "skill health check".

Remote control tmux sessions for interactive CLIs by sending keystrokes and scraping output. Use when you need to interact with a long-running process, REPL, or debugger. Triggers on: 'tmux', 'send keys', 'interact with pane', 'gdb', 'python repl'.

Find and assume roles from agent definitions in .github/agents/*.md files. Load agent context to guide specialized development workflows. Use when you need to discover and adopt specialized agent roles or personas. Triggers on: 'act as', 'persona', 'assume the role of', 'impersonate', 'who are the agents'.

Orchestrate the transition between GitHub Project iterations (e.g., ORG/PROJECT_NUMBER). Perform a "Sprint Stocktake" of open work and coordinate with the user to schedule the next 2-week iteration. Triggers on: 'sprint planning', 'iteration', 'sprint stocktake', 'plan the next sprint'.

Use this skill to identify and remove signs of AI-generated writing from text. Trigger this whenever asked to review, edit, or "humanize" a document to make it sound more natural, professional, and less like a chatbot. Essential for refining Architecture Decision Records (ADRs), policy documents, and executive communications by eliminating promotional language, filler, and repetitive patterns.

Use this skill whenever you need to create, read, edit, or manipulate Word documents (.docx files). Trigger this on any mention of 'Word doc', 'word document', '.docx', or requests to produce formatted deliverables like reports, memos, or letters. Use this for extracting text, inserting images, tracking changes, or converting content to Word format. Do NOT use for PDFs or spreadsheets.

Use this skill to structure and format Git commits. Trigger this whenever you are about to make a commit. This enforces the Atomic Commit Protocol (ACP), ensuring commits are self-contained, tested, documented, and conform to Conventional Commits format.

Find, cache, and refresh git repositories — both local projects and remote repos. Uses the gum CLI to discover existing checkouts under ~/*/ before cloning. Use when the user references a git repository (URL, org/repo, or local path) or when you need to locate a project on disk.

Git safety guardrails for AI agents. Read before ANY git operation. Ensures GPG signing is never disabled, editors never launch interactively, and prompts are suppressed. Triggers on any git commit, rebase, merge, or tag operation.

Use when the user mentions Jira issues (e.g., "PROJ-123"), asks about tickets, wants to create/view/update issues, check sprint status, or manage their Jira workflow. Triggers on keywords like "jira", "issue", "ticket", "sprint", "backlog", or issue key patterns.

Guide on creating and editing Mermaid charts safely. Use when asked to draw a diagram, flowchart, or architecture chart. Triggers on: 'mermaid', 'diagram', 'flowchart', 'visualize'.

Use this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text/tables from PDFs, combining or merging multiple PDFs into one, splitting PDFs apart, rotating pages, adding watermarks, creating new PDFs, filling PDF forms, encrypting/decrypting PDFs, extracting images, and OCR on scanned PDFs to make them searchable. If the user mentions a .pdf file or asks to produce one, use this skill.

Use this skill any time a .pptx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates, layouts, speaker notes, or comments. Trigger whenever the user mentions "deck," "slides," "presentation," or references a .pptx filename, regardless of what they plan to do with the content afterward. If a .pptx file needs to be opened, created, or touched, use this skill.

Use this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.

Configure and tune .codemap/config.json. Trigger this skill ONLY when codemap CLI output is overwhelmed by irrelevant files (like assets, generated code, or vendor folders), or when the prompt-submit hook explicitly suggests config-setup.

Systematic codebase exploration. Trigger this skill when you need to understand how a specific subsystem works, trace the downstream impact of a feature, find where a function is defined, or onboard to an entirely new area of the codebase.

Create and read cross-agent handoff artifacts. Trigger this skill when the conversation context window is full and you need to resume work in a new session, or when you are asked to pass your current progress to another autonomous agent (e.g., Claude, Cursor, Pi).

Strict safety protocol for modifying hub files. Trigger this skill immediately when you discover a file is imported by 3 or more other files, when the codemap hook surfaces 'hub-safety', or when making structural changes to core architecture files.

Analyze codebase structure, dependencies, changes, and multi-agent handoffs using codemap. Use this skill as the primary entry point when a task requires understanding project architecture, mapping dependencies, tracing logic, or generating cross-agent handoff artifacts. Always trigger this first when exploring unfamiliar code or when codemap output indicates noisy configurations.

Use this skill to orchestrate deep codebase exploration and complex structural refactoring by chaining Codemap, LSP Code Analysis, and AST Grep. Trigger this when asked to "safely refactor across the codebase", "trace this bug to its source", "explore the architecture of this subsystem", or whenever a task requires understanding the impact of a change before making it. This acts as a high-road strategy guide over basic grep/read commands.

Launch Kiro CLI as a headless sub-agent in a tmux pane and monitor its completion. Use when delegating a well-defined coding task to a Kiro agent subprocess. Triggers include "run kiro agent", "delegate to kiro", "kiro sub-agent", or any request to run kiro-cli headlessly.

Query Terraform Cloud (TFC) workspaces, runs, plans, logs, and team/project access via API. Use when inspecting TFC workspace state, checking run history, analyzing plan output, reviewing apply logs, managing team permissions, or diagnosing access issues in a Terraform Cloud organization.

AWS CLI safety guardrails and account/role management for AI agents. Read before running any AWS CLI command. Covers environment setup, credential checking, account switching, pagination, and account-guard patterns. Triggers on any aws CLI invocation, account switching, SSO login, or when setting up a shell pane for AWS work.

Diagnose and fix GitHub Actions workflow failures caused by Actions permissions policies — startup_failure with no jobs, third-party actions blocked, allowed_actions misconfiguration. Use when workflows fail immediately with no logs, or when actions from outside the repo owner are rejected.