armor

Deep-dive interview to flesh out a spec

Walkthrough branch changes with teaching-focused feedback

Write a concise recap to LOG.md

Review recent changes using specialized agents

Type design analyst — use when introducing new types, reviewing type changes in PRs, or refactoring existing types. Provides qualitative feedback and quantitative ratings on encapsulation, invariant expression, usefulness, and enforcement.

Error handling auditor — use when reviewing code changes for silent failures, inadequate error handling, inappropriate fallback behavior, or any code that could suppress errors. Invoke after completing work involving catch blocks, fallback logic, or error paths.

Test reviewer — use when reviewing tests in a diff, PR, or recently modified spec files. Drives a discussion by asking the questions that separate valuable tests from theater. Flags tautologies, over-mocking, framework-testing, vague names, and tests that would still pass if the prod code were gutted.

Expert second opinion — use when stuck on complex coding problems, architectural decisions, or technical challenges. Consults an external model for validation and synthesizes the response.

Jira ticket manager — use when creating, updating, or querying Jira tickets. Handles epics, stories, tasks, and bugs with proper formatting and project conventions.

Mutate one test file's assertions to verify they catch regressions. Receives a file path and diff context from the caller. Inverts assertions and confirms the tests fail.

Code simplifier — use after writing or modifying code. Simplifies for clarity, consistency, and maintainability while preserving all functionality. Focuses on recently modified code unless told otherwise.

REST API enforcer — use when creating, modifying, or reviewing REST API endpoints, routes, or HTTP layer code. Enforces Stripe-level standards for REST conventions, testing, naming, and consistency.

Verify new tests actually test new code. Catches false positives and over-mocking by gutting implementation in parallel isolated worktrees and checking if tests notice.

Web search and content extraction via Brave Search API. Use for searching documentation, facts, or any web content. Lightweight, no browser required.

Guidelines for authoring reusable UI components. Use when asked to create new frontend features, components, or design-adjacent tasks.

Diátaxis documentation framework — the four kinds of documentation (tutorials, how-to guides, reference, explanation) and how to apply the framework iteratively. This skill should be used whenever classifying documentation types, reviewing docs for boundary violations (a doc trying to be both a tutorial and reference), deciding how to structure new documentation, or applying the Diátaxis compass to improve existing docs. Also trigger when the user mentions "Diátaxis", "four kinds of documentation", "tutorial vs how-to", "documentation types", or asks whether a doc should be a tutorial, how-to, reference, or explanation.

Triage a documentation set — Confluence space, local folder, or any collection of docs — to find overlapping content, consolidation opportunities, staleness, and backlink candidates. Use this skill whenever the user asks to audit, triage, review, or health-check their documentation, knowledge base, wiki, or any collection of written content. Also trigger when the user mentions "doc overlap", "duplicate docs", "consolidate documentation", "backlinks", "documentation health", "stale docs", "knowledge base audit", "wiki cleanup", or wants to understand how their docs relate to each other. Works with Confluence, Google Docs, markdown files, Notion exports, or any text-based documentation source.

Always use when user asks to create, generate, draw, or design a diagram, flowchart, architecture diagram, ER diagram, sequence diagram, class diagram, network diagram, mockup, wireframe, or UI sketch, or mentions draw.io, drawio, drawoi, .drawio files, or diagram export to PNG/SVG/PDF.

Use this skill whenever you need to interface with Jira. For example, when the user asks to "make a ticket", "create a JIRA ticket", "file a ticket for this", "log this in JIRA", or mentions needing to interact with JIRA in any way. Also triggers on phrases like "turn this into a ticket" or "we should track this".

Verify WCAG 2.1.1 (Keyboard) compliance by testing fuzzy, property-based focus invariants against a live page. Use when auditing keyboard accessibility, testing focus management, validating tab order, or checking that interactive elements are reachable. Prefer Chrome MCP/devtools tools when available; CDP via curl is also supported. Framework-agnostic.

Detect unnecessary layout shifts (jank / CLS) in a web UI by instrumenting Chrome with PerformanceObserver while driving real interactions. Use whenever the user asks to check a UI for layout shifts, jank, jumpiness, content shifts, CLS, or Core Web Vitals — phrases like "check for layout shifts", "is it janky", "is anything jumping", "QA the UI", "any CLS issues". Also use proactively as part of any browser-based verification of a UI change: after taking screenshots or driving an interaction in the browser to confirm a feature works, run this pass to confirm the feature doesn't move already-painted content unexpectedly. The check is cheap (≈10 seconds in the same browser session) and catches a class of bug that static reading and casual screenshotting both miss.

Mine Figma comment threads to discover design patterns. Analyzes threads for repeated tensions, debates, exceptions, and recurring problems. Uses LLM reasoning to identify patterns from actual content rather than keyword matching. Requires Figma REST API access (FIGMA_PAT environment variable).

Use this skill when the user wants to set up a new TypeScript project or add standard tooling to an existing one. Triggers on phrases like "set up this repo", "initialize the project", "add linting", "scaffold this", "set up tooling", "configure typescript", or "sow".

Fetch content directly from URLs and return it as markdown, text, or raw HTML. Use when the user wants to read a webpage, inspect a URL, pull docs from a site, extract page content from a link, or fetch web content without opening a browser. Prefer this over browser automation when a normal HTTP fetch is enough. Supports GitHub blob URLs and sites that return LLM-optimized markdown for Accept: text/markdown.