The only memory skill that watches on its own. No database. No vectors. No manual saves. Just an LLM observer that compresses your conversations into prioritised notes, consolidates when they grow, and recovers anything missed. Five layers of redundancy, zero maintenance. ~$0.00/month (using free-tier models). While other memory skills ask you to remember to remember, this one just pays attention.
Generates 3-4 strategic talk angles with strength/weakness analysis, title options, CFP descriptions, and a peer feedback draft, then enforces a mandatory CHECKPOINT for user confirmation before scripting. Use when deciding how to frame a talk, preparing a CFP submission, or choosing between multiple narrative angles.
Diagnose and recover a Unity editor that is in Safe Mode, stuck compiling, stuck in a domain reload, or unreachable from MCP and CLI tools - detect the state with the Unity CLI, read only the compiler errors from Editor.log, fix the code, and restart cleanly. Use whenever every live Unity surface stops responding or the CLI reports Safe Mode.
Test-driven development skill for writing unit tests, generating test fixtures and mocks, analyzing coverage gaps, and guiding red-green-refactor workflows across Jest, Pytest, JUnit, Vitest, and Mocha. Use when the user asks to write tests, improve test coverage, practice TDD, generate mocks or stubs, or mentions testing frameworks like Jest, pytest, or JUnit.
Use when the user needs a verified, multi-model consensus answer for complex, high-stakes tasks. Invoke the PolyGnosis v3 adversarial debate pipeline: 3+ models solve independently with dynamic specializations and asymmetric tool allocation, a critic cross-reviews all solutions, severe bugs are logged to a Reflexion buffer, formal RRF+Borda ranking determines the winner, a Constitutional Quality Gate prevents synthesis regressions. Early Resolution circuit bypasses critique+scoring when solvers reach unanimous consensus. Do NOT use for simple queries or syntax fixes.
Comprehensive document creation, editing, and analysis with support for tracked changes, comments, formatting preservation, and text extraction. When Codex needs to work with professional documents (.docx files) for: (1) Creating new documents, (2) Modifying or editing content, (3) Working with tracked changes, (4) Adding comments, or any other document tasks
General-purpose agent for researching complex questions, searching for code, and executing multi-step tasks. Use when you need to perform comprehensive searches across codebase, find files that are not obvious in first few searches, or execute multi-step tasks requiring multiple tools and approaches.
iOS testing with XCTest/XCUITest on simulators and devices: layered strategy, determinism/flake control, device matrix selection, CI ergonomics, and repeatable simulator automation.
Maintain authoritative project progress across coding agents with leases, revision checks, evidence, migration and recovery. Use for explicit continuity work, or the project already contains `.relay/CURRENT.md`.
Track deals through every stage from lead to close. Manage pipeline stages, update deal status, forecast revenue, and identify bottlenecks in your sales process.
Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".
Resume any open PR — started by `om-auto-create-pr` or opened outside the pipeline. Claims it, resumes in an isolated worktree from the first unchecked step of its execution plan; a PR with no plan is adopted — the goal is reconstructed from its description, comments, review feedback, linked issues and diff, then executed. Usage - /om-auto-continue-pr <PR-number>
This skill should be used when the user says "configure hooks", "set up quality gates", "add PostToolUse hook", "set up permission hooks", "create hook configuration", "add typecheck hook", "set up secret-scan hook", or wants to configure event-driven hooks for their Claude Code project. Validates against the documented Claude Code event list — catches fictional event names before they end up in settings.json.