wheakerd

agents-architect

Design, initialize, audit, split, migrate, maintain, or validate a repository's AGENTS.md instruction system, routed .agents guidance, and repo-local .agents/skills that support that system. Use for active-instruction discovery, durable AGENTS updates, route ownership, oversized guidance, or an explicit effective-instructions, effective-instructions:preview, effective-instructions:refactor, effective-instructions:force, effective-instructions:reconcile, or effective-instructions:reconcile-preview mode. Do not use for packaged or installed plugin Skills, ordinary documentation, or general Codex usage optimization unless the requested change is specifically to AGENTS architecture.

confirm-external-action

Bind and verify a consequential external action. Use when the user explicitly asks to send, publish, invite, purchase, trade, delete, or change external app or account state and the exact actor, target, payload, disclosure, cost, or retry boundary matters. Do not use for read-only lookup, draft-only work, or local Git. Pair with reversible-system-change when the same request also has persistent rollback, data, service, or activation risk.

optimize-codex-usage

Reduce or diagnose Codex token, credit, and context consumption without weakening task quality or safety. Use only when the user explicitly asks to save Codex credits or tokens, optimize Skills, AGENTS.md, MCP, or context loading, explain unusually high Codex usage, or design a lower-consumption workflow. Do not use for ordinary coding, software performance work, or generic cost questions.

reversible-system-change

Plan, rehearse, or execute persistent system changes with verified rollback and postconditions. Use for an install, upgrade, deployment, migration, destructive retention action, or active-version promotion with rollback, data, service, or activation risk. Plans and non-mutating workflow rehearsals stay read-only; isolated restore rehearsals require exact rehearsal-write authority. Pair with confirm-external-action when the same request also causes a consequential external app or account effect. Do not use for ordinary source/configuration edits, Git operations, status queries, or conceptual explanations.

Code & Development Listed

review-axiom-task

Review an Axiom-guided task's routing, scope, authorization, actions, evidence, stops, and outcome. Use when the user asks what Axiom did; why it selected, allowed, or refused something; or to audit, criticize, appeal, or narrow that decision. Do not use for ordinary summaries, code review, execution, or retry.

Code & Development Listed

traceable-git-submit

Route explicit $traceable-git-submit; checkpoint/baseline, consolidation/recovery, hardened, multi-target, or history-replacing work; and a combined commit, tag, and push of an already-prepared plugin release. Ordinary named-remote non-force staging, commits, and pushes without a tag or those other triggers stay host-native; status, diff, and conceptual Git stay outside.

using-axiom

Route an Axiom plugin session to the smallest matching bundled skill. Use at startup, resume, or compaction, or when explicitly deciding whether Axiom applies; no-match requests continue normally.

agent-plugin-architect

Design, initialize, audit, migrate, maintain, or evaluate a packaged Codex or Claude Code plugin's shared Skills, route ownership, manifests, marketplace wrappers, hooks, and version-bound compatibility evidence. Use only for explicit packaged agent-plugin architecture work. Do not use for repository-local AGENTS.md or .agents/skills systems, ordinary source-code or documentation work merely because it is in a plugin repository, host installation, publication, deployment, or Git submission.