Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Cinderx Interpreter Case Analyze

ASecurity

Use when benchmark 未进入 CinderX JIT,需要分析解释执行阶段、函数形状和 gate 策略。

2 stars
0 votes
0 copies
0 views
Added 9/20/2026
toolspythonapiperformance

Works with

api

Security Analysis

A100/100

Scanned 9/20/2026

Install to Claude Code

$npx -y skills add sisibeloved/cpython-optimize-skill --skill cinderx-interpreter-case-analyze --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Cinderx Interpreter Case Analyze?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Cinderx Interpreter Case Analyze
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/sisibeloved-cinderx-interpreter-case-analyze/badge)](https://www.skillsdirectory.com/skills/sisibeloved-cinderx-interpreter-case-analyze)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
name: cinderx-interpreter-case-analyze
description: Use when benchmark 未进入 CinderX JIT,需要分析解释执行阶段、函数形状和 gate 策略。
---

# CinderX Interpreter Case Analyze

解释执行用例 / 非 JIT 用例不能硬套 HIR/LIR 结论。先证明 benchmark 本体未进入 CinderX JIT gate,或性能差距主要来自解释执行、gate 拒绝、运行时开销和 AutoJIT 分类决策,再做阶段分析。

## 前置证据

- 真实 pyperformance worker 命令和环境契约。
- `cinderx-jit-entry-check` 的 `entered_cinderx_jit=false`,或目标热函数未进入 gate 的证据。
- CPython JIT baseline、CinderX JIT 优化前、CinderX JIT 优化后的配对结果。
- perf、计数器、worker log、gate log、函数级调用次数或其它穿刺证据。

## 穿刺证据

穿刺证据必须把总耗时拆到可行动阶段,不能只写 speedup 结论。至少覆盖:

| 阶段 | 证据 |
|------|------|
| worker / benchmark 本体 | 真实 worker 命令、输入规模、warmup/loops、affinity、环境变量 |
| CPython JIT baseline | 基线结果、热点、JIT 状态、函数级证据 |
| CinderX JIT 优化前 | 优化前结果、gate / AutoJIT 状态、解释执行热点 |
| CinderX JIT 优化后 | 优化后结果、已改变阶段、残留热点 |
| 差距归因 | 哪些阶段贡献 gap,哪些阶段已被优化,哪些仍可优化 |

## 分阶段平铺表

输出一张平铺表,列必须包含:

| 阶段 | CPython JIT baseline | CinderX JIT 优化前 | gap | CinderX JIT 优化后 | 已优化量 | 剩余 gap | 证据 | 下一步 |
|------|----------------------|--------------------|-----|--------------------|----------|----------|------|--------|

要求:
- 每行只放一个阶段或一个可验证子阶段。
- `gap`、`已优化量`、`剩余 gap` 要有数值、比例或明确的定性证据。
- 不能把多个阶段揉成“解释器慢”一类结论。
- 没有证据的行标记为待补证据,不得当作根因。

## 函数形状表

分阶段平铺表之后,必须给函数形状表。基于 autojit 分类模型列出全量函数形状和 gate 策略,不只挑热点函数。

| 函数 | 模块/路径 | 调用/热度 | bytecode 形状 | 动态特性 | AutoJIT 分类 | gate 策略 | gate 结果 | 证据 |
|------|-----------|-----------|---------------|----------|--------------|-----------|-----------|------|

函数形状至少覆盖:
- loop、分支、异常处理、generator/coroutine、closure、globals、locals/freevars/cellvars。
- call 形态:Python call、C API、method/descriptor、vectorcall、dynamic dispatch。
- 数据形态:dict/list/tuple、attribute access、boxing/unboxing、type stability。
- AutoJIT 分类:可编译、待观察、拒绝、低收益、风险高、证据不足。
- gate 策略:阈值、白名单/黑名单、函数形状规则、平台规则、实验开关。

## 不进入 gate 的阶段详细拆解

只要有函数不进入 gate,就追加阶段详细拆解:

1. 发现:函数如何被发现,调用次数和热度是什么。
2. 分类:autojit 分类模型给出的函数形状和风险。
3. gate:命中哪条 gate 策略、阈值或拒绝规则。
4. 拒绝:不进入 gate 的直接原因。
5. fallback:解释执行 fallback 后落在哪些运行时阶段。
6. 代价:该拒绝对 CPython JIT baseline vs CinderX JIT 优化前 gap 的贡献。
7. 行动:调 gate、改函数形状支持、降低运行时开销或放弃优化的下一步。

## 输出

返回穿刺证据、分阶段平铺表、函数形状表、每个不进入 gate 函数的阶段详细拆解、已优化量、剩余 gap、风险和最小验证命令。

Attribution

sisibelovedsisibeloved
View sourceMore from sisibeloved →
SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Related Skills

ucoz-landing-skill

Playbook for creating and editing uCoz landing pages via MCP tools (`templates_tool`, `ftp_tool`, `modules_tool`). Use for tasks such as: "build a landing page", "update the homepage as a landing page", "create a promo page on the homepage", "add a lead form / menu / SEO to the homepage". Homepage: `page_list`, `page_get`; first publish — `page_update` with full `page_tmpl`; HTML edits after generation — `patch_template` (module_id=2, template_id=1), not `update_template`. Activate the mail f...

107 votes

Paperclip

Interact with the Paperclip control plane API for task coordination and governance. Use when checking assignments, updating issue status, posting comments, delegating work, managing routines, or calling Paperclip API endpoints.

805541 votes

Instantly Rdsthomas Mission Control

Instantly.ai cold email outreach API - manage campaigns, leads, accounts, and analytics. Use for cold email automation, lead management, campaign creation/monitoring, and email account warmup.

761 votes

Daw Music

Digital Audio Workstation usage, music composition, interactive music systems, and game audio implementation for immersive soundscapes.

761 votes

Caveman Compress

Compress natural language memory files (CLAUDE.md, todos, preferences) into caveman format to save input tokens. Preserves all technical substance, code, URLs, and structure. Compressed version overwrites the original file. Human-readable backup saved as FILE.original.md. Trigger: /caveman-compress FILEPATH or "compress memory file"

1023330 votes
View all in tools →