PaulKinlan

Discover team members, delegate tasks, and track progress to completion

Prepare structured meeting agendas and pre-reads from task board, artifacts, and team context

Classify, prioritize, and route incoming incidents based on severity, category, and affected components

Classify incoming requests and route to the appropriate specialist agent

How to communicate with other agents on the system. Use when you need to ask questions, share information, or coordinate.

Compare options against weighted criteria with scored matrix, sensitivity analysis, and quantified recommendation

Evaluate costs vs benefits of a proposed change, investment, or decision with quantified ROI

Analyze data files and produce summary statistics with real commands

Review and fix PR review bot findings on current PR, loop until resolved. Fetches unanswered bot comments, evaluates each finding, fixes real bugs, dismisses false positives, and replies to every comment with the outcome.

Expert code reviewer that checks code against project guidelines in CLAUDE.md with high precision to minimize false positives. Reviews for bugs, style violations, and code quality. Triggers: Before committing code, when reviewing changes, when checking code quality. Examples: - "Review my recent changes" -> reviews unstaged git diff against project guidelines - "Check if everything looks good" -> comprehensive code review - "Review this code before I commit" -> pre-commit quality check - "Check this PR" -> reviews all changes in the current PR

Analyzes code comments for accuracy, completeness, and long-term maintainability. Identifies misleading comments, comment rot, and documentation gaps. Triggers: After adding documentation, before finalizing a PR, when reviewing comments. Examples: - "Check if the comments are accurate" -> verifies comments match code behavior - "Review the documentation I added" -> analyzes new comments for quality - "Analyze comments for technical debt" -> finds outdated or misleading comments - "Are my docstrings correct?" -> validates documentation accuracy

Identifies silent failures, inadequate error handling, and inappropriate fallback behavior in code. Zero tolerance for errors that occur without proper logging and user feedback. Triggers: When reviewing error handling, checking for silent failures, analyzing catch blocks. Examples: - "Review the error handling" -> audits all error handling in recent changes - "Check for silent failures" -> hunts for swallowed errors and empty catch blocks - "Analyze catch blocks in this PR" -> reviews every try-catch for adequacy - "Are there any hidden failures?" -> finds errors that get silently ignored

Use this agent when code has been written or modified and needs to be simplified for clarity, consistency, and maintainability while preserving all functionality. This agent should be triggered automatically after completing a coding task or writing a logical chunk of code. It simplifies code by following project best practices while retaining all functionality. The agent focuses only on recently modified code unless instructed otherwise. Triggers: After writing or modifying code, when code needs simplification, when refactoring for clarity. Examples: - "Simplify this code" -> analyzes and simplifies recently modified code - "Make this clearer" -> improves readability while preserving behavior - "Refine this implementation" -> applies project standards and reduces complexity - "Clean up the code I just wrote" -> simplifies recent changes

Analyzes pull request test coverage quality and completeness. Focuses on behavioral coverage rather than line coverage, identifying critical gaps, test quality issues, and missing edge cases. Triggers: After writing tests, before creating a PR, when reviewing test coverage. Examples: - "Check if the tests are thorough" -> analyzes test coverage quality - "Review test coverage for this PR" -> maps tests to changed code - "Are there any critical test gaps?" -> identifies missing test scenarios - "Review my test quality" -> evaluates test resilience and patterns

Analyzes type design quality focusing on encapsulation, invariant expression, usefulness, and enforcement. Provides quantitative ratings (1-10) for each dimension. Triggers: When adding new types, reviewing type design in PRs, refactoring types. Examples: - "Review the UserAccount type design" -> analyzes type encapsulation and invariants - "Analyze type design in this PR" -> reviews all newly added types - "Check if this type has strong invariants" -> evaluates invariant enforcement - "How can I improve this type?" -> provides actionable type design suggestions

Fix all comments from the GitHub pull request for the current branch. Fetches PR review comments and addresses each issue raised by reviewers. Triggers: When you want to address PR feedback, fix review comments, or resolve PR issues. Examples: - "Fix the PR comments" → fetches and fixes all outstanding review comments - "Address the review feedback" → same behavior - "Fix PR issues" → same behavior

Drive the pi agent from, and reply to, Telegram, Discord, email, and webhooks via a CHAOS relay server. Use when the user wants to set up the relay bridge, connect a channel (Telegram bot, Discord bot, email, webhook), check for inbound messages from those channels, reply to a thread the agent received, or describe any "I want to message my agent / control pi from chat / email" need. This is the bridge between external channels and the pi agent.

Diagnose and recover from CHAOS relay problems — setup errors, "Failed to parse URL", messages not arriving, replies not sending, relay unreachable, channel pairing failures, corrupted config, WebSocket/transport issues, ECDSA signature errors, stale sessions. Use whenever the user reports the relay bridge is broken or behaving unexpectedly, when /chaos-relay status shows problems, or when an onboarding/registration call fails. The first action is always /chaos-relay doctor.

Run chrome-platform-showcase auto-research, conformance, and goal-setting workflows. Trigger when the user asks to run showcase auto research, critique demos, generate conformance suites, or turn critique open questions into demo work.

Audit a URL for modern web quality and (optionally) fix it. This is a FULLY AGENTIC audit. YOU (the model) gather multi-modal evidence about the page with the generic evidence primitives, decide for yourself which tools to run (Lighthouse, axe, your own ad-hoc static tests), reason over that evidence, and judge every principle, then emit a findings report. There are no hard-coded checks and no fast path. The principles are the spec, this skill is the method, and you supply all the intelligence. Use when asked to web-audit, UX-audit, uplift, modernise, or quality-audit a site.

Inventory the browser-runnable Hugging Face model universe, build comprehensive per-model demo pages (explainer + real in-browser inference + diagnostics + a wide use-case matrix), and validate them. Trigger when the user asks to run web-ai-showcase, refresh the model inventory, build/expand model pages, close the coverage gap, or validate the demos.

Public-alpha, stage-aware Chrome Developer Relations copilot for validating developer problems, rehearsing user and stakeholder perspectives, running friction and continuous-improvement loops, planning ecosystem adoption, protecting teams during contentious launches, and producing coherent enablement assets.

Audit chrome-platform-showcase coverage, build missing demos, and run critique, conformance, validation, and goal-setting workflows. Trigger when the user asks to run showcase auto research, fill demo-pending releases, critique demos, generate conformance suites, validate demos, or turn critique open questions into demo work.

Run a web-uplift audit of the top 1000 sites and produce structured results. Collects CDP evidence (screenshots, Lighthouse, axe, heap, layout, discoverability) and judges 17 quality principles per site. Outputs per-site JSON + merged SQLite DB.

Audit a URL for memory leaks. Generates (or reuses) a per-site interaction test plan, runs each scenario through the heap-snapshot methodology via chrome-devtools-mcp, analyzes snapshots with memlab, and writes a structured findings report. Use when asked to memory-audit, leak-check, or memory-trace a site.