AI Agent Skills

Records one architecture decision with alternatives, consequences and status; does not design the whole system.

Creates evidence-backed current or target architecture diagrams; not UI design.

Plans architecture migrations with compatibility, data safety, rollout and recovery; does not execute them.

Builds dependency-ordered delivery plans from requirements and repository evidence; read-only.

Plans risk-based test portfolios and acceptance evidence; does not write or execute tests.

Open one new native Codex task with a compact handoff, end this turn, and receive a single completion or blocker message back. Use when the user invokes $codex-relay and…

Coordinate authorized work across persistent AI agents and existing tasks. Use when the user asks to route work, hand off a task, check agent status, or manage several o…

Choose one legal next action in a browser, desktop, game or simulation. Supply fresh observed state and available actions. The host or simulator executes and checks the…

Locate, select, extract and verify evidence in documents or observed code inventories. Use for source-span extraction, passage reranking, claim checks or choosing code l…

Use for user-defined inbox, support-ticket, feedback or record classification and prioritization, especially bulk parallel judgments with sufficient per-record context.…

Apply before writing logic: choosing core types and data structures, sequencing scaffold-vs-feature work, asking what concurrent actors share. Get the data structures ri…

Apply when context is filling up: large outputs, long files, repeated reads, fan-out planning. Route bulk to subagents; keep summaries in the main thread, not raw payloa…

Apply when refactoring, evaluating diff size, or tempted to add abstractions, layers, or signal threading. Bias toward deletion and the smallest change that solves the p…

Apply when designing commands, lifecycle steps, or processing loops that run amid crashes, restarts, and retries. Converge to the same end state regardless of partial pr…

Apply when introducing a new internal API while old callers still exist. Migrate callers and delete the old API in the same wave instead of preserving compatibility laye…

Apply when reviewing or shaping code that's hard to trace. Count layers between question and answer, and hidden state in the reader's head; collapse one-caller wrappers…