本文へ移動
cccskills
無料GitHub で公開

proxy-dependent-python-service-diagnosis

Read-only diagnosis workflow for Python services whose external HTTP or WebSocket readiness differs between manual tests and service runtime behind a proxy.

インストール方法を見る

含まれるファイル(1)

  • SKILL.md4.3 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Proxy-Dependent Python Service Diagnosis

Purpose

Determine whether a Python service's external HTTP/WebSocket failures are caused by proxy environment propagation, client-library proxy behavior, proxy instability, stale readiness artifacts, or a different service entrypoint.

When to use

  • Manual network tests pass through a proxy but the service runtime fails.
  • Readiness checks report network unreachable, connection timeout, handshake timeout, or proxy-related errors.
  • HTTP proxy variables are configured but behavior is inconsistent.
  • A systemd-managed Python service behaves differently from same-user manual commands.

Safety rules

  • Default workflow is read-only.
  • Do not edit systemd units, wrappers, Python code, proxy config, or service environment without explicit approval.
  • Do not restart services or proxy daemons without approval.
  • Do not print proxy credentials, private endpoint URLs, tokens, cookies, or raw readiness payloads.
  • Redact concrete addresses and ports as <PROXY_URL>, <SERVER_IP>, <SERVICE_NAME>, <PROJECT_ROOT>, and <EXTERNAL_ENDPOINT>.

Read-only workflow

  1. Capture current failure class from safe summaries: HTTP failure, WebSocket failure, both, stale artifact, or mixed evidence.
  2. Inspect service metadata: user, working directory, command line, environment keys present, main PID, restart count, and unit entrypoint.
  3. Read the startup chain: systemd unit, wrapper script, virtual environment activation, module entrypoint, and any supervising loop. Summarize; do not expose secrets.
  4. Inspect the live child Python process environment through /proc/<pid>/environ, redacting values. Confirm whether proxy variables, virtualenv, path, and Python path reached the actual child process.
  5. Compare the live process command line with the manual command being tested.
  6. Inspect readiness/probe implementation names and client libraries involved: HTTP client, WebSocket client, custom wrappers, and whether the library honors environment proxy settings.
  7. Reproduce read-only probes in concept or in a controlled approved diagnostic session: same user, same virtualenv, same working directory, same Python path, with proxy and without proxy. If not approved to run probes, describe the test plan only.
  8. Check whether readiness artifacts are current: timestamp, PID/session correlation, and whether another process may be overwriting the same artifact.
  9. Separate conclusions for HTTP and WebSocket. Do not summarize one as proof for the other.
  10. Rank likely causes and identify the next safe diagnostic step.

Evidence to collect

  • Service metadata summary.
  • Live child process environment keys present, values redacted.
  • Entrypoint and wrapper chain summary.
  • HTTP client and WebSocket client/library versions if safely available.
  • Readiness artifact timestamp and status class.
  • Error class summaries for HTTP and WebSocket separately.
  • Manual-vs-service difference matrix.
  • Duplicate process or stale artifact indicators.

Stop conditions

  • Inspecting the process environment would expose secrets that cannot be safely redacted.
  • The next step requires code changes, service restart, proxy restart, or config edits.
  • Manual probes would hit write endpoints or trigger external side effects.
  • The current artifact timestamp is stale and no current evidence exists.

Optional user-approved actions

Only after explicit user approval:

  • Run same-user, same-venv diagnostic probes.
  • Add temporary debug instrumentation that records redacted proxy state.
  • Patch client configuration to use explicit proxy settings.
  • Restart the service or proxy daemon.
  • Disable duplicate manual launch chains that are overwriting artifacts.

Final report format

Summary: <proxy/runtime diagnosis>
Service: <SERVICE_NAME>
Current readiness: <status/timestamp/class>
Live process evidence: <env keys present, command summary>
HTTP path: <works/fails/unknown and evidence>
WebSocket path: <works/fails/unknown and evidence>
Artifact freshness: <current/stale/ambiguous>
Likely cause: <ranked hypothesis>
Next diagnostic step: <read-only or approval-required>
Needs approval: <specific action>

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Route CEO-first multi-window work into owner-led role loops. Use for 总控/CEO、架构/CTO、内容主编、角色派发与复用、模型路由、来源窗口回调、role-windows 台账、skill 命中统计或 multi-window loop engineering。

日本語の概要は準備中です。原文の説明を表示しています。

Dirtytrii/codex-skills112026年10月9日 更新

Read-only first workflow for diagnosing application incidents by collecting service, log, network, configuration, and functional evidence before proposing any fix.

日本語の概要は準備中です。原文の説明を表示しています。

Dirtytrii/codex-skills112026年10月9日 更新

Run user-authorized black-box web security tests and evidence-based reports. Use for 黑盒安全测试、渗透测试、数据泄露、未授权访问、公开 JS/API、登录爆破防护、CORS 或安全响应头检查。

日本語の概要は準備中です。原文の説明を表示しています。

Dirtytrii/codex-skills112026年10月9日 更新

Choose the narrowest reliable browser surface across Codex Browser, signed-in Chrome, Playwright CLI, and platform scripts. Use for browser automation, logged-in workflows, UI inspection, screenshots, form filling, publish previews, or frontend QA.

日本語の概要は準備中です。原文の説明を表示しています。

Dirtytrii/codex-skills112026年10月9日 更新

提议并执行 rubric 或 bucket 升级。两种模式:**完整 rubric bump**(最高风险动作,5 步强制 + 跨模型审核)和 **--bucket-only 轻量重校**(只换 bucket 边界,不动 rubric 公式)。**Phase 2 强制走 cheat-score-blind sub-agent 给校准池重打分**——不接受 self-scored fallback。触发词:"升级 rubric"/"bump rubric"/"更新公式"/"我想加一个维度"/"调整权重"/"重校桶"/"recalibrate bucket"。

日本語の概要は準備中です。原文の説明を表示しています。

Dirtytrii/codex-skills112026年10月9日 更新

初始化 cheat-on-content 内容校准工作区。仅在用户明确开始该工作流,或其子 skill 确认缺少 .cheat-state.json 时使用;不响应一般项目初始化、安装诊断或 Codex 首次会话。

日本語の概要は準備中です。原文の説明を表示しています。

Dirtytrii/codex-skills112026年10月9日 更新

Dirtytrii のスキルをすべて見る

このスキルの問題を報告する