/skills

Define intended behavior, review write scope and assess reversible decisions. Use when: discovery needs clarification or resumption before one complete slice; stop once actionable.

Implement changes, repairs or waves; return per-lane evidence. Use when: coding, service operations, reliability, delivery, incident recovery, resilience or toil is authorized.

Give advisory feedback on a plan, design or code. Use when: suggestions, tradeoffs or a second look are wanted. Not for acceptance or write scope; use Validate or Plan.

Freshly judge a finished change and its claims against original acceptance. Use when: acceptance verdict or independent proof is sought. Clarify generic checks or readiness first.

Coordinate authorized workers, prerequisites, isolated scopes and review capacity. Use when: dispatching, recovering or routing feedback. Not for implementation or judgment.

Find reviewed context, capture evidence or curate maintained claims. Use when: prior evidence can change an action, or learning is requested; no mandatory recall or lesson.

Write grounded docs, READMEs, repo instructions or continuity handoffs. Use when: these documents are requested; no reports as a routine completion ritual.

Clarify domain terms, bounded contexts and repository conventions. Use when: naming, rule ownership or Go and other language standards are unclear; avoid a broad survey.

Simplify structure, interfaces or responsibilities while preserving behavior. Use when: a focused refactor is requested; feature changes need their own intent.

Trace code or test a recurring pattern to answer one cited question. Use when: uncertainty needs evidence. Not for external feature teardowns; use reverse-engineer.

Tear down an authorized competitor repo, binary or product into a feature inventory and adoption choices. Use when: comparing an external system; local questions go to Research.

Review code or scan for security vulnerabilities, secrets, dependencies and prompt risks. Use when: concrete exposure needs assessment; never silently change policy.

Create, adapt, consolidate or repair skill packages and projections. Use when: authoring guidance, descriptions or structure; Skill Eval measures behavioral benefit.

Measure whether a skill helps a named task or needs revision or removal. Use when: a bounded routing or coding evaluation is requested; conformance alone cannot show benefit.

Write behavioral tests, practice TDD or inspect important coverage gaps. Use when: test design or missing proof needs work; running an existing suite needs no skill.

Compare independent views on a consequential or contested decision. Use when: the caller selects multiple judges; evidence resolves disagreement, not voting.

Draft or lint a bounded persistent goal above a bead graph of RPI experiments. Use when: this goal workflow is explicitly selected; shaping a single change belongs to Plan.

Generate evidenced options or challenge an idea. Use when: deciding what to build or comparing alternatives; exploration does not authorize implementation.

Analyze outcomes or an interim cutoff. Use when: a postmortem is explicitly requested; consumes available judgment, never gates code acceptance or requires a lesson.

Challenge a rollout plan with one fresh judge before implementation; identify what could make it fail. Not for finished-code judgment. Triggers: "one judge", "challenge this plan".

Audit claimed state, goals or native status. Use when: a claim audit or snapshot is requested. Clarify advice versus acceptance for ambiguous checking or readiness requests.

Apply the outcome-to-judgment charter. Use when: the caller explicitly selects RPI; ordinary coding, delegation and native goals do not require this workflow.

Switch coding-agent accounts and verify runtime identity. Use when: the caller requests an account change; never rotate automatically to evade a quota.

Coordinate selected writers with Agent Mail messages and advisory file reservations. Use when: this adapter is requested; mail does not own tracker status.

Dispatch independent tasks to parallel workers or selected persistent roles. Use when: delegation is authorized with disjoint scopes; execution does not validate output.

Run a supplied task in AGY Antigravity and collect its result. Use when: the caller selects AGY; never a fallback for native coding.

Search agent session logs and cited episodes with CASS. Use when: past prompts, decisions or failures may answer a question; repeated text is not a proven lesson.

Configure Claude Code hooks and narrow enforcement guards. Use when: the caller requests hook installation, repair or policy changes; a hook is not required to use other skills.

Run one prompt through headless Codex and capture its result. Use when: requesting a single noninteractive Codex process. Not for worker batches or retries.

Diagnose a Destructive Command Guard block or configure its rules. Use when: DCG rejected an operation or policy work is requested; never disguise commands to bypass it.

Find and load guidance with the meta_skill search engine. Use when: searching a skill corpus; CASS owns past sessions and Skill Builder owns package authoring.

Operate selected NTM agent panes and inspect native state. Use when: persistent tmux roles are requested; pane liveness and prompt delivery are not validation.

Offload one build through RCH or diagnose its remote compiler. Use when: remote compilation is selected; report errors without creating a retry controller.

Inspect disk pressure with SBH and perform an authorized recovery action. Use when: storage diagnosis or SBH recovery is requested; inspection does not authorize deletion.

Operate the Agentic Coding Flywheel through its native workflow. Use when: the caller explicitly selects this factory; convergence and closed work do not prove semantic acceptance.

Operate Gas City through its Mayor, registry packs and native run state. Use when: the caller explicitly selects Gas City; factory completion does not replace independent judgment.