Claude Skills by machenjie

Dispatch-only.

For analysis/task/review agents using a Professional Skill on models, RAG, agents, evaluation, or safety; not for work without AI decision impact.

Use when an installed-client change has a confirmed Android target and changes Android platform behavior.

For analysis/task/review agents using a Professional Skill on batch, stream, warehouse, lineage, replay, or data quality; not for transactions without distributed-data impact.

Use when a selected Professional decision materially depends on confirmed cloud-provider policy, ownership, or behavior.

Use when an installed-client change has a confirmed shared framework and concrete registered platform targets.

Use when an installed-client change has a confirmed iOS or iPadOS target and changes Apple mobile platform behavior.

For analysis/task/review agents using a Professional Skill on devices, firmware, edge protocols, updates, or safety; not for work without device, firmware, or physical impact.

Use when a confirmed Linux graphical desktop target changes platform behavior.

For analysis/task/review agents using a Professional Skill on kernels, drivers, ABI/FFI, memory, concurrency, or syscalls; not for work without a native boundary.

Use when an installed-client change has a confirmed macOS target and changes Apple desktop platform behavior.

For analysis, task, or review agents using a Professional Skill on payments, trading, ledgers, settlement, refunds, or reconciliation; not for non-financial or Web3-only work.

For analysis/task/review agents using a Professional Skill on wallets, signing, smart contracts, chain transactions, or custody; not for work with no chain behavior.

Use when a confirmed Windows target changes desktop or service platform behavior.

Use with named Profiles for one narrow engineering decision; do not use without the registered trigger or as the primary task owner.

`analysis-agent`/`task-agent`/`review-agent`: use when completion needs measurable acceptance evidence; skip when no task-local acceptance standard must be defined.

`analysis-agent`/`task-agent`/`review-agent`: use when UI semantics, keyboard/focus, announcements, contrast, motion, alternatives, or scaling changes; skip non-UI work.

Use when execution-evidence source, freshness, scope, reproducibility, reuse, or contradictions are uncertain; skip when current claim-matched evidence is established.

`analysis-agent`/`task-agent`/`review-agent`: classify command risk when targets, mutation, recovery, effects, or authorization are unresolved; skip when facts are current.

`analysis-agent`/`task-agent`/`review-agent`: use when scale, access, complexity, memory, streaming, or skew drives an algorithm/data-structure choice; skip when no choice exists.

`analysis-agent`/`task-agent`/`review-agent`: use when endpoints, payloads, errors, auth, pagination, idempotency, or compatibility need a contract; skip when APIs are unchanged.

`analysis-agent`/`task-agent`/`review-agent`: use when architecture rules need lint, dependency, type, or CI enforcement; skip when no enforcement decision is required.

`analysis-agent`: use when selecting monolith, modular, layered, service, event-driven, or hybrid architecture; skip when no style decision is required.

`analysis-agent`: use when architecture options, constraints, material consequences, and exit costs need an evidence-backed comparison; skip when no task-local tradeoff exists.

`task-agent`/`review-agent`: use when jobs, workers, queues, schedules, retries, cancellation, or status visibility need design; skip when no async-job decision is required.

Use for audit coverage, identity/time fidelity, tamper evidence, retention, export, or custody decisions. Skip event emission, telemetry design, privacy purpose, and legal claims.

`analysis-agent`/`task-agent`/`review-agent`: use for authenticated-subject authority, propagation, or authorization handoff changes; skip credential lifecycle or policy-only work.

Use with analysis-agent, task-agent, or review-agent for task-local authentication lifecycle and recovery risk. Do not use without that decision or as task owner.

`task-agent`/`review-agent`: use when protected state, restore objectives, dependency order, or recovery evidence changes; skip backup-job-only work with no recovery decision.

`analysis-agent`/`task-agent`/`review-agent`: use when build graphs, codegen, caches, toolchains, or artifact reproducibility change; skip when build behavior is unaffected.

`analysis-agent`/`task-agent`/`review-agent`: use when changes may alter money, permissions, tenants, or audit rules; skip presentation-only work with unchanged invariants.

`analysis-agent`/`task-agent`: use when extracting policies, calculations, constraints, or invariants from scattered layers; skip when no business-rule decision is needed.

Use with task-agent or review-agent for task-local cache scope, freshness, invalidation, and source-load risk. Do not use without a cache decision or as task owner.

`analysis-agent`/`task-agent`/`review-agent`: use when CI/CD checks, artifacts, deployment gates, IaC, or Helm behavior changes; skip when pipelines are unaffected.

`analysis-agent`/`task-agent`/`review-agent`: use when deleting dead code, flags, fallbacks, compatibility, APIs, or generated remnants; skip when no cleanup decision exists.

`analysis-agent`/`task-agent`/`review-agent`: use when CLI, TUI, or daemon commands, flags, config, streams, exit codes, signals, or dry-run behavior change; skip unrelated work.

`analysis-agent`/`task-agent`/`review-agent`: use for installed-client lifecycle, permissions, activation, installation, device, or accessibility tests; skip general quality work.

`analysis-agent`/`task-agent`/`review-agent`: use for installed-client lifecycle, termination, restoration, or identity-reset decisions; skip static UI and platform-API-only work.

`analysis-agent`/`task-agent`/`review-agent`: use when changed flow, naming, or navigation hides obligations; skip placement, API-contract, language, or performance decisions.

`analysis-agent`/`task-agent`/`review-agent`: use when variables, expressions, or statements can alter local semantics; skip structure, API, language, or runtime ownership.

`analysis-agent`/`task-agent`/`review-agent`: use when code needs correctness, security, performance, maintainability, or hallucinated-API review; skip when no review is needed.

`analysis-agent`/`task-agent`/`review-agent`: primary-Skill-selected for races, locks, optimistic conflicts, or worker overlap; never task owner; skip without concurrency impact.

`analysis-agent`/`task-agent`/`review-agent`: use when typed config, defaults, flags, modes, rollout, rollback, or cleanup changes; skip when runtime policy is unaffected.

`analysis-agent`/`task-agent`/`review-agent`: use when API, SDK, schema, event, package, CLI, or export changes may affect consumers; skip when no consumer contract can change.

`task-agent`/`review-agent`: use when image layers, build context, runtime user, secrets, health checks, shutdown, or provenance change; skip when container behavior is unaffected.

`analysis-agent`/`task-agent`/`review-agent`: use for provider-consumer compatibility of APIs, events, schemas, or behavior; skip without independent consumer/version risk.

`task-agent`: use when implementing routes, request parsing, validation handoff, auth context, responses, or status/error mapping; skip work without a controller/API boundary.

`analysis-agent`/`task-agent`/`review-agent`: use when C/C++ changes cross ownership, UB, ABI/FFI, exception, concurrency, or target-build boundaries; skip tool-only work.

Use for cryptographic construction, envelopes, key lifecycle, compromise, destruction, or agility. Skip secrets, authentication, provider APIs, custom primitives, and compliance.

`analysis-agent`/`task-agent`/`review-agent`: use for C# async, resource, type, DI, trimming/AOT, loading, interop, UI, or deployment decisions.