跳到正文
cccskills
免费在 GitHub 公开

proxelar

Intercept, inspect, and modify HTTP/HTTPS traffic via a Rust MITM proxy with Lua scripting hooks. Supports API debugging, LLM call capture for evals, endpoint mocking, CORS/auth header injection, latency simulation, and telemetry blocking. Triggers on 'proxy', 'intercept traffic', 'inspect requests', 'mock API', 'capture API calls', 'MITM', 'network debug', 'HTTP inspection'.

中文简介正在准备中。当前显示原文说明。

查看安装方法

包含的文件(10)

  • SKILL.md4.7 KB
  • README.md5.8 KB
  • references/lua-scripting-api.md2.9 KB
  • scripts/block_telemetry.lua1.0 KB
  • scripts/capture_for_eval.lua2.0 KB
  • scripts/frametime_bench.js1.7 KB
  • scripts/inject_cors.lua1.6 KB
  • scripts/latency_inject.lua981 B
  • scripts/log_to_jsonl.lua1.5 KB
  • scripts/mock_api.lua1.3 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

Proxelar

Rust-based MITM proxy for intercepting and transforming HTTP/HTTPS traffic via Lua scripting. Provides TUI, terminal, and web GUI interfaces. Installed via brew install proxelar.

Category

Infrastructure Operations

Modes

ModeCommandWhen to use
Forward proxyproxelarIntercept outbound traffic from any app configured to use 127.0.0.1:8080
Reverse proxyproxelar -m reverse --target http://localhost:3000Inspect traffic to a local dev server without configuring the client
Web GUIproxelar -i guiAgent-visible network inspector at http://localhost:8081 — screenshot for visual feedback
Scriptedproxelar --script path/to/hook.luaAutomated request/response transformation via Lua hooks

Combine modes: proxelar -m reverse --target http://localhost:3000 -i gui --script scripts/log_to_jsonl.lua

CLI Reference

FlagDescriptionDefault
-i, --interfaceterminal, tui, guitui
-m, --modeforward, reverseforward
-p, --portListening port8080
-b, --addrBind address127.0.0.1
-t, --targetUpstream target (required for reverse mode)--
--gui-portWeb GUI port8081
--ca-dirCA certificate directory~/.proxelar
-s, --scriptLua script for request/response hooks--

Lua Scripting

Define on_request and/or on_response hooks:

function on_request(request)
    -- request.method, request.url, request.headers, request.body
    -- Return request to forward (modified or not)
    -- Return response table to short-circuit: { status = 403, headers = {}, body = "Blocked" }
    -- Return nil to pass through unchanged
end

function on_response(request, response)
    -- response.status, response.headers, response.body
    -- Return response (modified or not), or nil to pass through
end

Bundled Lua Scripts

Run any script with proxelar --script ~/.claude/skills/proxelar/scripts/<name>.lua

ScriptPurpose
log_to_jsonl.luaLog request/response pairs as JSONL to ~/.proxelar/traffic.jsonl
capture_for_eval.luaCapture LLM API calls (Anthropic/OpenAI/Gemini) as structured JSONL for eval pipelines
mock_api.luaTemplate for mocking API endpoints — edit the mocks table
inject_cors.luaAdd permissive CORS headers to all responses
latency_inject.luaAdd configurable artificial latency to matching URLs
block_telemetry.luaBlock analytics and telemetry domains

Agent Integration

Agent-Browser Bridge

Route agent-browser traffic through Proxelar for combined page + network visibility:

# Terminal 1: start proxelar in reverse mode
proxelar -m reverse --target http://localhost:3000 -i gui --script ~/.claude/skills/proxelar/scripts/log_to_jsonl.lua &

# Terminal 2: navigate through proxelar, then snapshot
agent-browser --proxy "http://127.0.0.1:8080" open http://127.0.0.1:8080 && agent-browser snapshot

Agent Visual Feedback via Web GUI

Start Proxelar with -i gui, then screenshot http://localhost:8081 via Claude-in-Chrome or agent-browser to give the agent network-layer visibility into what the application is doing.

Frametime + Network Correlation

Inject scripts/frametime_bench.js into a running page via mcp__claude-in-chrome__javascript_tool or agent-browser eval. Read console output with pattern p50 to get percentile data. Run alongside latency_inject.lua to correlate network latency with rendering jank.

CA Certificate

First run auto-generates ~/.proxelar/proxelar-ca.pem. Trust it once:

sudo security add-trusted-cert -d -r trustRoot -k /Library/Keychains/System.keychain ~/.proxelar/proxelar-ca.pem

Or visit http://proxel.ar through the proxy for guided install.

Gotchas

  • Check port 8080 is free before starting: lsof -i :8080
  • Reverse mode requires --target — CLI errors immediately if omitted
  • Lua scripts are loaded once at startup; edit requires restart
  • Web GUI binds to 127.0.0.1:8081 by default; use --addr 0.0.0.0 for remote access
  • WebSocket traffic passes through but Lua hooks only fire on HTTP request/response, not individual WS frames
  • TUI and Web GUI support interactive request inspection, editing, and dropping — not just passive viewing

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

Search academic papers, build literature reviews, and synthesize research findings — combines Exa search (publication category, domain and date filters) with arxiv-mcp-server for paper discovery, download, and deep analysis. Triggers on academic paper, literature review, research synthesis, arxiv, find papers, scholarly search.

中文简介正在准备中。当前显示原文说明。

tdimino/claude-code-minoan412026年10月20日 更新

Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.

中文简介正在准备中。当前显示原文说明。

tdimino/claude-code-minoan412026年10月20日 更新

This skill should be used when creating, auditing, or maintaining AGENTS.md files and Codex CLI configuration for any project—including initializing AGENTS.md for cross-agent compatibility (Codex, Cursor, Copilot, Devin, Jules, Amp, Gemini CLI), generating config.toml or .rules files, scaffolding .agents/skills/, converting CLAUDE.md to AGENTS.md, or auditing existing agent configs for bloat and staleness. Complementary to codex-orchestrator (which executes subagents; this skill creates the config files they consume).

中文简介正在准备中。当前显示原文说明。

tdimino/claude-code-minoan412026年10月20日 更新

Academic research skill for Biblical Hebrew, Semitic linguistics, cuneiform studies, and comparative Ancient Near Eastern research. Provides Sefaria API for Hebrew Bible, CDLI/ORACC for cuneiform databases, and web discovery via Omnisearch, Exa, Firecrawl, and Obscura for finding scholarly sources across JSTOR, Perseus, Persée, Google Scholar, and academia.edu. Triggers on Hebrew quotes, cuneiform, Sefaria, ANE research, Minoan, search for scholarship, find papers, literature review, scholarly search, academic search, Genesis/Tehom, Ugaritic, Talmudic sources, extract from PDF, OCR academic.

中文简介正在准备中。当前显示原文说明。

tdimino/claude-code-minoan412026年10月20日 更新

Build comprehensive ARCHITECTURE.md files following matklad's canonical guidelines — bird's-eye views, ASCII/Mermaid diagrams, codemaps, invariants, and layer boundaries. Triggers on document the architecture, create ARCHITECTURE.md, map this codebase, architectural overview.

中文简介正在准备中。当前显示原文说明。

tdimino/claude-code-minoan412026年10月20日 更新

Generate physically-based atmospheric scattering shaders — sky domes, planetary atmospheres, LUT-optimized pipelines, depth-aware post-processing. Four modes — sky-dome, atmosphere-post, planet, lut. Triggers on atmospheric scattering, sky shader, sunset rendering, planet atmosphere, Rayleigh scattering, Mie scattering, volumetric sky, sky dome, atmosphere post-processing, aerial perspective, transmittance LUT, sky rendering, realistic sky, planetary rendering.

中文简介正在准备中。当前显示原文说明。

tdimino/claude-code-minoan412026年10月20日 更新

查看 tdimino 的全部技能

报告此技能的问题