Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/sisibeloved/cpython-optimize-skill/cinderx-evidence-tablenpx skills add sisibeloved/cpython-optimize-skill --skill cinderx-evidence-tablegit clone --depth 1 https://github.com/sisibeloved/cpython-optimize-skillWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/sisibeloved/cpython-optimize-skill/cinderx-evidence-table)<a href="https://agentmods.dev/skills/sisibeloved/cpython-optimize-skill/cinderx-evidence-table"><img src="https://agentmods.dev/badge/skills/sisibeloved/cpython-optimize-skill/cinderx-evidence-table.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00085 | $0.01634 |
| Opus 5 | $0.00043 | $0.00817 |
| Sonnet 5 | $0.00017 | $0.00327 |
| Haiku 4.5 | $0.00009 | $0.00163 |
Grade A, and why
cinderx-evidence-table scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CinderX Evidence Table
证据表是单用例深钻的唯一产物。一张表 9 行（E1–E9），每行三段式：What（必贴客观证据）/ Verdict（一句结论）/ Gate（证据足够硬条件）。任一段不满足 Gate，证据表标记 evidence_gap 或 verdict_missing，流程停下来补证据，不跳过。
证据表本身不负责"先钻哪个用例"——那是 workflow 层的选例动作。证据表从 E1 起假设用例已选定。
E1–E9 结构
| # | 证据段 | What（必贴） | Verdict | Gate（证据闭环判据） |
|---|---|---|---|---|
| E1 | 性能基线 | 双平台 wall clock、speedup、置信区间、样本数、运行环境句柄 | 该用例是否存在显著且稳定的平台差距 | 差距 > 噪声带（默认 3%）且置信区间不含 0 |
| E2 | 用例画像 | 用例内容描述、autojit 函数形状（签名、热点循环结构） | 该用例的优化价值定位（compute-bound? memory-bound? branchy?） | 函数形状与 perf 热点吻合 |
| E3 | HIR 分布 | HIR 节点分布表（按 pass 或 node kind 统计）、双平台 HIR 是否同构 | 差异是否源于编译期（HIR 层） | 双平台 HIR 形态已比对，差异点已定位到节点级 |
| E4 | LIR + wall clock | LIR/机器码段、调用计数、wall clock 拆解（该段占总时间比） | 差异是否源于 codegen（LIR 层） | wall clock 拆解能与 LIR 段对齐，热点段已锁定 |
| E5 | 差异点 | bb 维度整体差异 或 单条机器码级差异，双平台逐指令对齐 | 具体差异指令/序列是什么 | 双平台指令已逐条对齐，差异序列 ≤ N 条可枚举 |
| E6 | 根因下钻 | 跨层证据链：LIR→机器码→ISA/指令集→微架构→硬件 | 表面原因之下的真实根因 | 根因已落到 ISA/微架构/硬件某一层，且非表面原因 |
| E7 | 优化方向 | 由根因推出的优化策略（不是猜测） | 该方向能否直接解决 E6 根因 | 策略 ↔ 根因一一对应，无悬空假设 |
| E8 | 穿刺数据 | A/B 验证（baseline=原形态，candidate=优化形态），wall clock + 必要 PMU | 优化是否真的消除差异、是否带来收益 | 穿刺隔离满足 cinderx-ab-run-slot，结果可重复 |
| E9 | 优化价值判定 | 净收益（穿刺收益 − 实现成本 − 风险）、通用性、是否进备选 | 该优化项是否进入备选优化项清单 | 收益与成本均有数据支撑，无纯定性结论 |
E6 根因层级必带工具证据
通用必带（所有平台）：perf stat 的 cycles / instructions / cache-references / cache-misses / branch-misses，双平台同口径采集。
| 根因层 | 必带工具输出（缺则 evidence_gap） |
|---|---|
| 机器码 | objdump -d 反汇编，双平台对齐 |
| ISA/指令集 | 指令选择差异、SIMD/原子/barrier 形态（对照 cinderx-isa-microarch-compare 矩阵） |
| 微架构 | perf stat 关键计数器（cycles/instructions/cache-miss/branch-miss）、perf top |
| 硬件 | PMU 计数器、必要时的 uops.info / llvm-mca 吞吐估算、cache/TLB 行为 |
指令级采样机制（SPE/IBS）处理原则
指令级精确采样在两平台是不同机制、都有硬性开启条件，且具体可用性随平台/内核/权限而变。本节只规定处理原则，不绑定任何特定平台的可用性结论——具体某平台能否拿到 SPE/IBS，由运行时探测决定，不预设。
-
采样可用性是证据链前置判据，非默认前提。E6 开头先做探测，探测结果本身作为 E6 第一条证据记录：
- ARM 侧：
perf list | grep arm_spe，并检查/sys/bus/event_source/devices/arm_spe_*/是否存在。为空则 SPE 证据不可得，退化为 PMU 计数器 +perf top。 - x86 侧：检查
/sys/bus/event_source/devices/ibs_op/是否存在，并确认权限——IBS 无用户/内核过滤能力，需CAP_SYS_ADMIN或CAP_PERFMON。无特权则 IBS 证据不可得，退化为 top-down metric group。 - 探测结果明确"本平台能拿到什么层级的证据、缺什么"。
- ARM 侧：
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 58 lines · 85 tokens per session scan A 024cdc7d1f26
cinderx-evidence-table is a skill published in the GitHub repository sisibeloved/cpython-optimize-skill (2 stars, last pushed 7d ago), licensed MIT. It adds 85 tokens to every session and 1,634 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
debug-optimize-lcp
Guides debugging and optimizing Largest Contentful Paint (LCP) using Chrome DevTools MCP tools. Use this skill whenever the user asks about LCP performance, slow page loads, Core Web Vitals optimization, or wants to understand why their page's main content takes too long to appear. Also use when the user mentions…
systematic-debugging
Use when debugging a failing test, build error, or runtime issue that isn't immediately obvious. Guides a 4-phase root cause analysis instead of random fix attempts.
diagnose
Trace from a reproduced symptom to the source code that causes it. Pin the specific file and approximate line, rate confidence in the cause and clarity of the fix independently, and always propose a concrete fix.
azsdk-common-pipeline-analysis
Analyze Azure SDK CI/CD pipeline failures into a structured diagnosis, and define the required output format. Load this skill before calling azsdkanalyzepipeline, which returns raw failure data that this skill interprets and formats. USE FOR: "pipeline failed", "build failure", "CI check failing", "tests failing in…
repro-admin
Reproduce an EmDash admin UI bug. Attach a container, start the demo dev server, drive the admin with agent-browser using the dev-bypass session, and capture the reproduction as screenshots plus a replayable transcript.
log-error-digest
Analyze log files to troubleshoot errors, identify peak error periods, and produce error clustering, frequency statistics, and time distribution reports. Supports JSON, syslog, and Nginx formats with automatic detection. Use when a user uploads a .log file and asks to analyze errors, find patterns, debug issues, or…