菜单
Profile flameview itself under high sample rates and heavy folded inputs, identify parser/viewer hotspots, and validate improvements with benches/tests. Use when optimizing flameview performance or verifying regressions after profiler-related changes.
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。
Profile Rust binaries on Linux using cargo flamegraph/perf custom events, keep folded stacks, and inspect them with flameview. Use when you need repeatable CPU/custom-event profiling with minimal dependencies, or when converting allocation profiles into flameview-ingestible folded stacks.
Drive interactive flameview terminal sessions and verify UX behavior (columns, filters, navigation, auto-expand). Use when a task requires launching flameview on folded stacks, exercising key bindings, reproducing TUI bugs, or capturing manual UX validation steps.
Create and maintain snapshot-style regression tests for flameview interactive UI by replaying key sequences and asserting textual screen snapshots with Insta. Use when UI layout, key handling, filters, or heuristics change and you need deterministic regression coverage.
基于 SOC 职业分类
| name | flameview-optimize-self |
| description | Profile flameview itself under high sample rates and heavy folded inputs, identify parser/viewer hotspots, and validate improvements with benches/tests. Use when optimizing flameview performance or verifying regressions after profiler-related changes. |
Use this skill to run a repeatable self-optimization loop for flameview:
Generate a large folded stack file:
<repo-root>/.codex/skills/flameview-optimize-self/scripts/make_heavy_folded.sh \
<repo-root> 160
Profile flameview on that file at a high sample rate:
<repo-root>/.codex/skills/flameview-optimize-self/scripts/profile_self_high_rate.sh \
<repo-root> \
/tmp/flameview-self.folded \
999
Run quick parser benchmarks for regression checks:
<repo-root>/.codex/skills/flameview-optimize-self/scripts/bench_parser_fast.sh \
<repo-root>
flameview --summarize to inspect generated folded outputs deterministically.flameview-tui-driver.references/optimization-loop.md for a concise optimization checklist.