treygoff24

Use when completing tasks, implementing major features, or before merging to verify work meets requirements - runs review in a forked code-reviewer agent

Activate autonomous loop mode for persistent development sessions. **CRITICAL:** Always activate this AFTER completing spec and implementation plan, BEFORE beginning implementation. This is what makes autonomous builds actually autonomous—the Stop hook enforces completion criteria.

Use when creating or developing, before writing code or implementation plans - refines rough ideas into fully-formed designs through collaborative questioning, alternative exploration, and incremental validation. Don't use during clear 'mechanical' processes

Apply systematic root cause analysis and debugging methodologies to diagnose and fix bugs, test failures, and unexpected behavior. Use when encountering production issues, investigating test failures, diagnosing performance problems, tracing error sources through call stacks, analyzing logs and stack traces, reproducing inconsistent bugs, debugging race conditions, investigating memory leaks, or applying scientific method to problem-solving before proposing fixes.

Activate orchestrator mode. You coordinate specialists, you don't implement. MAXIMIZE parallelism with task-builder. MAXIMIZE quality with code-reviewer, codex, gemini. Speed + Quality through delegation, not through doing work yourself.

Execute ONE task in parallel with other task-builders. ALWAYS SPAWN MULTIPLE for independent tasks. ORCHESTRATOR: You coordinate, you don't implement. Spawn N task-builders for N unblocked tasks. Maximum parallelism = maximum speed = maximum quality (each task gets focused attention).

Use when starting feature work that needs isolation from current workspace or before executing implementation plans - creates isolated git worktrees with smart directory selection and safety verification

Use when design is complete and you need detailed implementation tasks for engineers with zero codebase context - creates comprehensive implementation plans with exact file paths, complete code examples, and verification steps assuming engineer has minimal domain knowledge

Use when operating Memorum from the memoryd CLI — orienting in a new session, searching and reading memories, recording notes and governed memories, superseding or forgetting, revealing encrypted content, checking daemon health, or importing/backing up prior Claude Code / Codex CLI memory. Covers the full agent operating loop and the JSON envelope + exit-code contract every command follows.

Use the elv CLI for ElevenLabs audio and voice work from the command line: generate speech (TTS), transcribe audio (STT), create sound effects or music, clone, convert, or isolate voices, dub video/audio, and manage ElevenLabs voices, models, conversational agents, history, and usage. elv also invokes ANY of the ~320 ElevenLabs API operations by id, plus raw REST and WebSocket calls. It is agent-first: every command is non-interactive and prints exactly one JSON envelope to stdout, with binary and large output written to disk. Reach for elv whenever the user wants ElevenLabs text-to-speech, speech-to-text, voice cloning, sound effects, music, dubbing, or any ElevenLabs REST/WebSocket call.

Build a standalone HTML explainer/decision-doc for the Próspera Radar project in the house style, token-efficiently. Use when Trey asks for an HTML explainer, walkthrough, decision doc, architecture brief, or visual brief. Author only the body fragment; the shared CSS is inlined by assemble.py so no stylesheet is ever regenerated.

Check repository occupancy before edits, find which local agents are running, claim a workspace before writing, release claims afterward, and diagnose fleet sources. Use before modifying a repo or launching concurrent agent work.

Check repository occupancy before edits, find which local agents are running, claim a workspace before writing, release claims afterward, and diagnose fleet sources. Use before modifying a repo or launching concurrent agent work.

Implements and verifies Rust parser contract, adapter integration, and parser-facing reliability work.

Finalizes parser mission docs, reliability surfaces, and full quality-gate verification.

Implements and verifies normalization, events, mechanics, report, schema, and docs work for the parser mission.

AI coaching for Rocket League using local replay analysis. Use when the user asks about their Rocket League performance, replays, improvement, or coaching. Covers analysis, comparison, weakness detection, and practice recommendations.

A worker that implements features and writes tests in the Specgate Rust codebase.

Codebase health scanner and technical debt tracker. Use when the user asks about code quality, technical debt, dead code, large files, god classes, duplicate functions, code smells, naming issues, import cycles, or coupling problems. Also use when asked for a health score, what to fix next, or to create a cleanup plan. Supports 28 languages.

Use the elv CLI for ElevenLabs audio and voice work from the command line: generate speech (TTS), transcribe audio (STT), create sound effects or music, clone, convert, or isolate voices, dub video/audio, edit Dubbing Project transcripts, and manage ElevenLabs voices, account-visible models, conversational agents, workspaces, history, and usage. elv invokes all 338 operations compiled from its pinned ElevenLabs OpenAPI document, plus raw REST and protocol-aware WebSocket calls. It is agent-first: every command is non-interactive and prints exactly one JSON envelope to stdout, with binary and large output written to disk. Reach for elv whenever the user wants ElevenLabs text-to-speech, speech-to-text, voice cloning, sound effects, music, dubbing, or any ElevenLabs REST/WebSocket call.

Audit and improve how well a CLI works when an AI agent is the primary user — JSON envelopes, exit-code contracts, error pedagogy, intent inference, determinism, non-TTY discipline. Use when checking the agent-fitness of elv (or any CLI we build), hardening robot/JSON output, or fixing agent-hostile error messages. Ponytail-compatible: the smallest fix that clears the bar, ranked by leverage, no quota — zero findings is a valid result.

Cut and ship a receipts release end to end — version bump, changelog, contract-surface sync, tag, cargo-dist CI, crates.io publish, and channel verification. Use when the user says cut a release, ship a release, publish a new version, or tag vX.Y.Z.

Get source-verified answers with the receipts CLI — every claim returns with a source URL, quote, and verdict. Use when you need a citable answer to a factual question, when a claim is about to go into a document and must be ground-truthed first, or when the user asks to fact-check something. Spends real money per question (~$0.05–0.35).

Use when the user says /autonomous-loop, use autonomous-loop, work autonomously until done, prevent early exit, keep going until quality gates pass, or asks Codex not to stop until a task list is complete. Enables the autonomous-loop CLI, creates and continuously maintains the Codex task plan with update_plan, implements the task, runs trusted gates from .codex/autoloop.project.json, and does not finish until every plan item is completed and verification gates pass.

Use the exa-agent CLI for Exa search, contents, answer, context, similar, Websets, monitors, and agent workflows.

Use when operating Memorum from the memoryd CLI — orienting in a new session, searching and reading memories, recording notes and governed memories, superseding or forgetting, revealing encrypted content, checking daemon health, or importing/backing up prior Claude Code / Codex CLI memory. Covers the full agent operating loop and the JSON envelope + exit-code contract every command follows.