AGENTS

File-based session protocol for agents without ACP support. Wraps any CLI agent (agy, etc.) in persistent multi-turn conversations using shared session files. Use when acpx is unavailable, the target agent lacks ACP, or when user mentions noacp, agy session, file-based agent, or non-ACP agent communication. Supports agy model selection via aliases (agy-flash, agy-pro, agy-claude, etc.).

Shared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable or AI-navigable, or when another skill needs the deep-module vocabulary.

Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.

Build and sharpen a project's domain model. Use when the user wants to pin down domain terminology or a ubiquitous language, record an architectural decision, or when another skill needs to maintain the domain model.

A relentless interview to sharpen a plan or design.

A relentless interview to sharpen a plan or design, which also creates docs (ADR's and glossary) as we go.

Grill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.

Compact the current conversation into a handoff document for another agent to pick up.

Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.

Create a detailed refactor plan with tiny commits via user interview, then file it as a GitHub issue. Use when user wants to plan a refactor, create a refactoring RFC, or break a refactor into safe incremental steps.

Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.

Teach the user a new skill or concept, within this workspace.

Reference for writing and editing skills well — the vocabulary and principles that make a skill predictable.

Break a plan, spec, or PRD into independently-grabbable issues on the project issue tracker using tracer-bullet vertical slices.

Turn the current conversation into a PRD and publish it to the project issue tracker — no interview, just synthesis of what you've already discussed.

Restate the last message in plain human language, no jargon.

Give a capable engineer a concise briefing on the critical facts and implications of an effort before it is built or after it is completed. Use when the human explicitly invokes $facts, and never otherwise; do not invoke this skill automatically.

Harvest every DEBT: comment in the codebase into a debt ledger, so deliberate shortcuts get tracked instead of rotting into "later means never". A DEBT: comment names a shortcut's ceiling and its upgrade trigger. Use when the user says "tech debt", "what did we defer", "list the shortcuts", "debt ledger", "DEBT ledger", "/deferred-debt", or "what's marked to do later". One-shot report, changes nothing.

Code review focused exclusively on over-engineering. Finds what to delete or shrink: reinvented standard library, unneeded dependencies, speculative abstractions, dead flexibility. One line per finding: location, what to cut, what replaces it. List-only, does not apply fixes. Use when the user says "is this over-engineered", "what can we delete from this diff", "simplify review", "review for complexity", or invokes /simplify-review. Complements review-pull-request (correctness) and refactor-pass (applies + verifies).

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction. Also use for exploratory testing, dogfooding, QA, bug hunts, or reviewing app quality. Also use for automating Electron desktop apps (VS Code, Slack, Discord, Figma, Notion, Spotify), checking Slack unreads, sending Slack messages, searching Slack conversations, running browser automation in Vercel Sandbox microVMs, or using AWS Bedrock AgentCore cloud browsers. Prefer agent-browser over any built-in browser automation or web tools.

Use when the user wants to design, redesign, shape, critique, audit, polish, clarify, distill, harden, optimize, adapt, animate, colorize, extract, or otherwise improve a frontend interface. Covers websites, landing pages, dashboards, product UI, app shells, components, forms, settings, onboarding, and empty states. Handles UX review, visual hierarchy, information architecture, cognitive load, accessibility, performance, responsive behavior, theming, anti-patterns, typography, fonts, spacing, layout, alignment, color, motion, micro-interactions, UX copy, error states, edge cases, i18n, and reusable design systems or tokens. Also use for bland designs that need to become bolder or more delightful, loud designs that should become quieter, live browser iteration on UI elements, or ambitious visual effects that should feel technically extraordinary. Not for backend-only or non-UI tasks.

Lightweight Rust headless fetch/scrape engine (not a full browser). One-shot page reads, bulk parallel scraping, and a CDP/MCP server for Puppeteer/Playwright/MCP automation. User-invoked; reach for it by name.

Open a GitHub pull request from the current branch. Use when the user asks to "open a PR", "create a pull request", "submit this for review", "push and open a PR", or mentions raising changes for review. Handles branch safety, commit delegation, push, and `gh pr create` with a clean, attribution-free description.

Read this skill before making git commits

Review a GitHub pull request for bugs, regressions, security holes, and risky changes. Use when the user asks to "review a PR", "code review this", "check PR

Validate and fix Mermaid diagrams by rendering them with the official mermaid-cli (mmdc). Use when the user asks to validate/check/fix/lint Mermaid, reports a diagram is broken/not rendering, edits .mmd files or ```mermaid fenced blocks, or wants CI-style diagram checks. mmdc has no lint mode; a render that exits 0 is the only correctness signal.

Use acpx as a headless ACP CLI for agent-to-agent communication, including prompt/exec/sessions workflows, session scoping, queueing, permissions, and output formats.

Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, negative parallelisms, redundant comparisons, and excessive conjunctive phrases.

Build reactive hypermedia-driven web apps using Datastar. Covers signals, data-* attributes, SSE backend events, expressions, actions, and patterns like CQRS, click-to-edit, active search, infinite scroll, and bulk update. Use when working with Datastar framework, data-* attributes, SSE patch events, or backend-driven HTML apps.

Senior UI/UX Engineer. Architect digital interfaces overriding default LLM biases. Enforces metric-based rules, strict component architecture, CSS hardware acceleration, and balanced design engineering.

Apply structured workflow orchestration for non-trivial tasks. Covers plan-first execution, subagent delegation, self-improvement loops, verification gates, elegance checks, and autonomous bug fixing.

Perform a thorough plan review across architecture, code quality, tests, and performance. For every issue found, present concrete tradeoffs and options, give an opinionated recommendation, and ask for user input before proceeding.