菜单
Use when planning, implementing, reviewing, or interpreting TSZ performance work, including benchmark regressions, cache/residency changes, timing claims, perf counters, hotspot investigations, OOM/timeout/stack-overflow blockers, or optimization PR evidence.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | tsz-performance-engineering |
| description | Use when planning, implementing, reviewing, or interpreting TSZ performance work, including benchmark regressions, cache/residency changes, timing claims, perf counters, hotspot investigations, OOM/timeout/stack-overflow blockers, or optimization PR evidence. |
Use for performance, cache/residency, OOM/timeout/stack-overflow, counters, benchmark regressions, and timing claims.
AGENTS.md, docs/plan/ROADMAP.md, and relevant PR/issues.TypeId comparisons.references/perf-mistakes.md before adding/widening caches, changing
residency, or claiming speedups.Use narrow, reproducible commands; wrap heavy runs.
python3 scripts/perf/cache-visibility-report.py --json
python3 scripts/perf/visited-clone-report.py --json
python3 scripts/perf/debug-print-report.py --json
python3 scripts/perf/migration_callsite_counts.py --json
python3 scripts/perf/query-perf-counters.py --json <artifact> --baseline <baseline>
scripts/safe-run.sh ./scripts/bench/perf-hotspots.sh --quick --json-file /tmp/hotspots.json
scripts/safe-run.sh ./scripts/bench/bench-vs-tsgo.sh --filter '<row>' --json-file /tmp/bench.json
Use focused compile guard or cargo nextest run -E 'test(...)' when shortest.
Do not run full conformance, emit, fourslash, or broad project suites locally.
NodeIndex and
cross-interner TypeId.any, target/module/options, request
scope, cycle/fuel, file/session generation?PR/comment packet: Project Corpus Impact, invariant, exact commands/CI, green
row timing only, RSS/failure-class evidence for runtime blockers, counter
deltas, noise/caveats, AgentName.
Preserve TSZ architecture boundaries while changing checker, solver, binder, emitter, LSP, WASM, or CLI code. Use when planning semantic fixes, moving logic across query boundaries, handling architecture guard failures, ratcheting boundary debt, or reviewing changes for solver-first and no hardcoding rules.
Drive TSZ GitHub PRs through review, CI, and the merge queue. Use when checking PR status, debugging failing GitHub Actions, deciding whether a PR is WIP, marking draft PRs ready, enqueueing or verifying the queue, landing a PR, or interpreting TSZ CI gates.
Triage and maintain TSZ diagnostic conformance. Use when investigating conformance regressions, accepted-regression drift, fingerprint-only failures, issue creation from conformance data, or focused parity fixes that must preserve the conformance gate.
Prevent TSZ machines and worktrees from running out of disk while preserving useful build caches. Use before creating worktrees, running large builds/tests, diagnosing low disk, cleaning artifacts, deciding whether to delete caches, or recovering space without slowing future Rust/TypeScript builds.
Audit TSZ repo iteration speed and quality. Use when asked to deep-dive repo health, identify avoidable delays, improve agent workflows, reduce CI/PR churn, add guardrails, choose skills to install or create, or turn recurring TSZ process failures into focused tooling/docs PRs.
Create, refresh, and publish TSZ pull requests with correct coordination state. Use when opening a draft PR, updating a PR body, applying canonical agent labels, changing WIP/draft/ready state, acknowledging review, handing off readiness to a manager agent, or making sure a TSZ PR satisfies AgentName, Project Corpus Impact, verification, and overlap rules.