Agent skill

Codegraph

Run CodeGraph CLI commands through npx for local codebase indexing, semantic search, context building, and impact analysis.

MITAuto-check passedAgent Workflows

Install Codegraph

skills CLI

$ npx skills add aiguicai/MCP-Gateway --skill codegraph -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI

$ gh skill install aiguicai/MCP-Gateway codegraph --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy

$ git clone --depth 1 https://github.com/aiguicai/MCP-Gateway.git skills-src && mkdir -p .claude/skills && cp -r skills-src/mcp-gateway/crates/gateway-http/builtin-skills/codegraph .claude/skills/codegraph && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Paste into Claude Code

Install the "codegraph" agent skill from https://github.com/aiguicai/MCP-Gateway/tree/main/mcp-gateway/crates/gateway-http/builtin-skills/codegraph into .claude/skills/codegraph/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "codegraph", then confirm the skill loads.

Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

$skill-installer

$skill-installer install https://github.com/aiguicai/MCP-Gateway/tree/main/mcp-gateway/crates/gateway-http/builtin-skills/codegraph

Type this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.

skills CLI

$ npx skills add aiguicai/MCP-Gateway --skill codegraph -a codex

Project install goes to .agents/skills/; add -g for ~/.codex/skills/.

GitHub CLI

$ gh skill install aiguicai/MCP-Gateway codegraph --agent codex

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/aiguicai/MCP-Gateway.git skills-src && mkdir -p .agents/skills && cp -r skills-src/mcp-gateway/crates/gateway-http/builtin-skills/codegraph .agents/skills/codegraph && rm -rf skills-src

Use ~/.agents/skills/ instead of .agents/skills for a personal install.

Codex skills documentation · loads skills from .agents/skills/

Paste into Codex

Install the "codegraph" agent skill from https://github.com/aiguicai/MCP-Gateway/tree/main/mcp-gateway/crates/gateway-http/builtin-skills/codegraph into .agents/skills/codegraph/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "codegraph", then confirm the skill loads.

Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add aiguicai/MCP-Gateway --skill codegraph -a cursor

Project install goes to .agents/skills/; add -g for ~/.cursor/skills/.

GitHub CLI

$ gh skill install aiguicai/MCP-Gateway codegraph --agent cursor

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/aiguicai/MCP-Gateway.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/mcp-gateway/crates/gateway-http/builtin-skills/codegraph .cursor/skills/codegraph && rm -rf skills-src

Use ~/.cursor/skills/ instead of .cursor/skills for a personal install.

Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/

Paste into Cursor

Install the "codegraph" agent skill from https://github.com/aiguicai/MCP-Gateway/tree/main/mcp-gateway/crates/gateway-http/builtin-skills/codegraph into .cursor/skills/codegraph/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "codegraph", then confirm the skill loads.

Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

gemini skills install

$ gemini skills install https://github.com/aiguicai/MCP-Gateway.git --path mcp-gateway/crates/gateway-http/builtin-skills/codegraph

--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.

skills CLI

$ npx skills add aiguicai/MCP-Gateway --skill codegraph -a gemini-cli

Project install goes to .agents/skills/; add -g for ~/.gemini/skills/.

GitHub CLI

$ gh skill install aiguicai/MCP-Gateway codegraph --agent gemini-cli

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/aiguicai/MCP-Gateway.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/mcp-gateway/crates/gateway-http/builtin-skills/codegraph .gemini/skills/codegraph && rm -rf skills-src

Use ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.

Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/

Paste into Gemini CLI

Install the "codegraph" agent skill from https://github.com/aiguicai/MCP-Gateway/tree/main/mcp-gateway/crates/gateway-http/builtin-skills/codegraph into .gemini/skills/codegraph/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "codegraph", then confirm the skill loads.

Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

GitHub CLI

$ gh skill install aiguicai/MCP-Gateway codegraph

Installs for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).

skills CLI

$ npx skills add aiguicai/MCP-Gateway --skill codegraph -a github-copilot

Project install goes to .agents/skills/; add -g for ~/.copilot/skills/.

Manual copy

$ git clone --depth 1 https://github.com/aiguicai/MCP-Gateway.git skills-src && mkdir -p .github/skills && cp -r skills-src/mcp-gateway/crates/gateway-http/builtin-skills/codegraph .github/skills/codegraph && rm -rf skills-src

Use ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.

GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/

Paste into GitHub Copilot

Install the "codegraph" agent skill from https://github.com/aiguicai/MCP-Gateway/tree/main/mcp-gateway/crates/gateway-http/builtin-skills/codegraph into .github/skills/codegraph/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "codegraph", then confirm the skill loads.

GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

skills CLI

$ npx skills add aiguicai/MCP-Gateway --skill codegraph -a opencode

OpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.

GitHub CLI

$ gh skill install aiguicai/MCP-Gateway codegraph --agent opencode

Project scope by default (.agents/skills/); add --scope user for a personal install.

Manual copy

$ git clone --depth 1 https://github.com/aiguicai/MCP-Gateway.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/mcp-gateway/crates/gateway-http/builtin-skills/codegraph .opencode/skills/codegraph && rm -rf skills-src

Use ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.

OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/

Paste into OpenCode

Install the "codegraph" agent skill from https://github.com/aiguicai/MCP-Gateway/tree/main/mcp-gateway/crates/gateway-http/builtin-skills/codegraph into .opencode/skills/codegraph/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "codegraph", then confirm the skill loads.

OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.

Facts

Skill name: codegraph
GitHub stars: 158
Token cost: ~1.3k tokens
SKILL.md length: 422 words
Files: 1
Skills in repo: 5
Repo updated: May 27, 2026
First seen: Oct 7, 2026
Licence: MIT

At a glance

Run CodeGraph CLI commands through npx for local codebase indexing, semantic search, context building, and impact analysis.

Works in 4 steps: Do NOT use shell_command to run… → Executing commands (skillToken required) → The gateway runs CodeGraph through npx.… → …
Agent Workflows work in your project
SKILL.md covers IMPORTANT: How to Use This Tool, Prerequisites, Allowed Commands and Typical Workflow, plus 1 more section
Calls npx

What it does

Codegraph is an agent skill from aiguicai/MCP-Gateway. Run CodeGraph CLI commands through npx for local codebase indexing, semantic search, context building, and impact analysis. First read the complete builtin://codegraph/SKILL.md to get skillToken; this SKILL.md read does not require skillToken. Calls without the correct skillToken will fail and must be retried.

Its SKILL.md is about 1.3k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Agent Workflows. It works with Model Context Protocol and npm. The repository describes itself as: Unify multiple MCP Servers & Skills into a single gateway — with proxy forwarding, authentication, and a management API. Typical use: expose local stdio servers as remote MCP… The licence is MIT.

When your agent uses it

Agent Workflows work in your project

Example prompts

“/codegraph”

Requirements

Node.js

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

Do NOT use shell_command to run CodeGraph commands. The gateway blocks
Executing commands (skillToken required)
The gateway runs CodeGraph through npx. You write codegraph ..., and
This is NOT the CodeGraph MCP server integration. Do not run

What it can do on your machine

Read from SKILL.md and the folder at commit 9cfc330. It shows what the files ask for, not the result of running them.

Tool permissions
Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.
From allowed-tools in the SKILL.md frontmatter.
Runs code
Shell commands in SKILL.md call:
- npx
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use npx, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Codegraph loads about 1.3k tokens when it runs. Until then it costs about 80 tokens; SKILL.md has 422 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it: ~80
When it runs · the whole SKILL.md, loaded when a task matches: ~1.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from aiguicai/MCP-Gateway at commit 9cfc330, republished under its MIT licence (© aiguicai). 422 words, ~1,259 tokens.

Download SKILL.mdSave it as .claude/skills/codegraph/SKILL.md (or your agent's skills folder).

name: codegraph
description: Run CodeGraph CLI commands through npx for local codebase indexing, semantic search, context building, and impact analysis. First read the complete builtin://codegraph/SKILL.md to get skillToken; this SKILL.md read does not require skillToken. Calls without the correct skillToken will fail and must be retried.

CodeGraph

CodeGraph builds a local semantic graph of a codebase and exposes CLI commands for indexing, symbol search, task context, file structure, call relationships, impact analysis, and affected-test discovery.

Use this tool when a repository has or should have a .codegraph/ index and the task needs structural code intelligence: "where is this symbol", "how does this area work", "who calls this", "what might break if I change it", or "which tests are affected by these files".

IMPORTANT: How to Use This Tool

This is a dedicated built-in tool named codegraph. You MUST follow these rules:

Do NOT use shell_command to run CodeGraph commands. The gateway blocks attempts to run codegraph or npx -y @colbymchenry/codegraph through shell_command when this tool is enabled. Always use this codegraph tool.
Executing commands (skillToken required):
- Every exec value MUST start with the codegraph prefix.
- You MUST include the skillToken returned from reading this document.
- Example: codegraph({"exec": "codegraph status", "cwd": "D:/project", "skillToken": "<token>"})
The gateway runs CodeGraph through npx. You write codegraph ..., and the gateway executes npx -y @colbymchenry/codegraph ... directly without a shell wrapper.
This is NOT the CodeGraph MCP server integration. Do not run codegraph serve --mcp. This built-in tool only supports selected CLI commands. Upstream MCP-only tools such as codegraph_trace, codegraph_explore, and codegraph_node are not available through this built-in tool.

Prerequisites

Node.js/npm must be installed so npx is available. The upstream npm package currently requires a supported Node runtime, and the first call may download or refresh the @colbymchenry/codegraph npm package cache.

Always set cwd to the project root or a directory inside the configured allowed directories. CodeGraph stores its local SQLite index under .codegraph/.

Show full SKILL.md (147 more words)Show less

Allowed Commands

Use these commands through the codegraph tool:

bash

codegraph --version
codegraph help [subcommand]
codegraph init [path] -i
codegraph index [path]
codegraph sync [path]
codegraph status [path]
codegraph query <search> [--kind <kind>] [--limit <n>] [--json]
codegraph files [path] [--format <format>] [--filter <glob>] [--max-depth <n>] [--json]
codegraph context <task> [--max-nodes <n>] [--max-code <n>] [--no-code] [--format markdown|json]
codegraph callers <symbol> [--limit <n>] [--json]
codegraph callees <symbol> [--limit <n>] [--json]
codegraph impact <symbol> [--depth <n>] [--json]
codegraph affected <files...> [--depth <n>] [--filter <glob>] [--json] [--quiet]

Blocked commands:

bash

codegraph serve --mcp
codegraph install
codegraph uninstall
codegraph uninit
codegraph unlock

Do not use the no-argument codegraph installer through this tool.

Typical Workflow

Check whether the project is initialized:

bash

codegraph({"exec": "codegraph status", "cwd": "D:/project", "skillToken": "<token>"})

Initialize and index a project when needed:

bash

codegraph({"exec": "codegraph init -i", "cwd": "D:/project", "skillToken": "<token>"})

Search for a symbol or inspect a broad area:

bash

codegraph({"exec": "codegraph query AuthService", "cwd": "D:/project", "skillToken": "<token>"})
codegraph({"exec": "codegraph context \"how authentication middleware works\" --max-nodes 30", "cwd": "D:/project", "skillToken": "<token>"})

Inspect call relationships:

bash

codegraph({"exec": "codegraph callers AuthService --json", "cwd": "D:/project", "skillToken": "<token>"})
codegraph({"exec": "codegraph callees AuthService --limit 30", "cwd": "D:/project", "skillToken": "<token>"})
codegraph({"exec": "codegraph impact AuthService --depth 2", "cwd": "D:/project", "skillToken": "<token>"})

Find affected tests or files:

bash

codegraph({"exec": "codegraph affected src/auth.ts", "cwd": "D:/project", "skillToken": "<token>"})

Notes

Prefer codegraph status before assuming the project is indexed.
Use codegraph sync after large file changes if the index looks stale.
Use codegraph context first for architecture or "how does this work" questions; use query for exact symbol lookup; use callers, callees, or impact for change planning.
Prefer --json when you need structured output for follow-up reasoning.
Do not use shell redirection, pipes, or affected --stdin with this tool. If you need affected tests, pass file paths as arguments. If you need to persist output, use multi_edit_file for files and read_file for inspection.

Files

Just SKILL.md in mcp-gateway/crates/gateway-http/builtin-skills/codegraph of aiguicai/MCP-Gateway.

Open the folder on GitHubat commit 9cfc330

Compare with similar skills

Codegraph next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Codegraph compared with similar skills
Skill	Stars	Used in	Tokens	Auto-check	Licence	Repo updated
Codegraph this skillaiguicai/MCP-Gateway	158	—	~1.3k	Automated safety check: Pass	MIT	4 mo ago
MCP Server BuildershareAI-lab/learn-claude-code	78k	5 repos	~1.2k	Automated safety check: Pass	MIT	9 days ago
Orca Run Replayiflytek/skillhub	5.2k	4 repos	~3k	Automated safety check: Pass	Apache-2.0	today
Yahoo Finance2gadicc/yahoo-finance2	801	—	~1.8k	Automated safety check: Pass	MIT	4 days ago
Workflow Schema Tuningbreaking-brake/cc-wf-studio	5.4k	—	~1.3k	Automated safety check: Pass	Custom licence	yesterday
Paseo Plugin Buildergetpaseo/paseo	20k	—	~9.4k	Automated safety check: Pass	Custom licence	today

Similar skills

MCP Server Builder
shareAI-lab/learn-claude-code
Walks through building MCP servers in Python or TypeScript that expose tools, resources and prompts to Claude, with templates, registration and testing.
78k GitHub starsUsed in 5 repos~1.2k tokens
Agent WorkflowsAuto-check passed
Orca Run Replay
iflytek/skillhub
Answers questions about a past agent run from its recording, using causal graphs and replay, instead of reconstructing events from memory.
5.2k GitHub starsUsed in 4 repos~3k tokens
Agent WorkflowsAuto-check passed
Yahoo Finance2
gadicc/yahoo-finance2
A skill your agent uses when building with the yahoo-finance2 TypeScript/Deno/npm library, using its Yahoo Finance data modules, CLI, or MCP server, or contributing to the yahoo-finance2 repository…
801 GitHub stars~1.8k tokensUpdated 4 days ago
Agent WorkflowsAuto-check passed
Workflow Schema Tuning
breaking-brake/cc-wf-studio
Guides edits to cc-wf-studio's workflow schema so AI agents generate better workflows, treating schema text as prompt engineering rather than validation.
5.4k GitHub stars~1.3k tokensUpdated yesterday
Agent WorkflowsAuto-check passed
Paseo Plugin Builder
getpaseo/paseo
Builds, installs and troubleshoots trusted local Paseo plugins, from lifecycle hooks and slash commands to screens, themes and timeline renderers.
20k GitHub stars~9.4k tokensUpdated today
Agent WorkflowsAuto-check passed
Create Harness
ruvnet/metaharness
Scaffold your own focused AI agent harness — pick host (Claude Code, Codex, pi.dev, Hermes), template, agents, skills, and ship a npm-publishable harness with its own npx CLI.
690 GitHub stars~777 tokensUpdated today
Agent WorkflowsAuto-check passed

More from aiguicai/MCP-Gateway

Chrome Cdp
aiguicai/MCP-Gateway
Browser automation and debugging through bundled Chrome DevTools Protocol.
158 GitHub stars~2.1k tokensUpdated 4 mo ago
Auto-check passed
Task Planning
aiguicai/MCP-Gateway
Use this bundled skill to create, update, and clear the in-memory todo state required before real bundled tool calls when task planning is enabled.
158 GitHub stars~1.1k tokensUpdated 4 mo ago
Auto-check passed
Chat Plus Adapter Debugger
aiguicai/MCP-Gateway
为 Chat Plus 新框架编写、修复、审查站点适配脚本。先读取完整 builtin://chat-plus-adapter-debugger/SKILL.md 获取 skillToken；这次读取 SKILL.md 不需要 skillToken。不要用正则或局部读取只抓 token。后续调用不带或带错 skillToken 会报错并必须重试，所以正式调试前先拿到正确 token。
158 GitHub stars~3.8k tokensUpdated 4 mo ago
Auto-check passed
Officecli
aiguicai/MCP-Gateway
Create, analyze, proofread, and modify Office documents (.docx, .xlsx, .pptx) using the officecli CLI tool.
158 GitHub stars~7.1k tokensUpdated 4 mo ago
Auto-check passed

Works with

Questions about Codegraph

What does Codegraph do?

Run CodeGraph CLI commands through npx for local codebase indexing, semantic search, context building, and impact analysis. Codegraph is an agent skill from aiguicai/MCP-Gateway. Run CodeGraph CLI commands through npx for local codebase indexing, semantic search, context building, and impact analysis.

When should I use Codegraph?

Codegraph fits situations like: agent Workflows work in your project.

How do I install Codegraph in Claude Code?

Run `npx skills add aiguicai/MCP-Gateway --skill codegraph -a claude-code`. Or copy the skill folder (mcp-gateway/crates/gateway-http/builtin-skills/codegraph in aiguicai/MCP-Gateway) into .claude/skills/codegraph in your project. Claude Code loads it when a task matches its description.

How do I install Codegraph in Codex?

Run `npx skills add aiguicai/MCP-Gateway --skill codegraph -a codex`. Or copy the skill folder (mcp-gateway/crates/gateway-http/builtin-skills/codegraph in aiguicai/MCP-Gateway) into .agents/skills/codegraph in your project. Codex loads it when a task matches its description.

Can I use Codegraph in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add aiguicai/MCP-Gateway --skill codegraph -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/codegraph, .gemini/skills/codegraph, .github/skills/codegraph and .opencode/skills/codegraph in your project.

What does Codegraph need to run?

Going by SKILL.md and its folder, Codegraph needs the command-line tools its instructions call (npx). Our summary lists: Node.js.

Does Codegraph access the network?

SKILL.md contains no URLs. Its commands use npx, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Codegraph safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Codegraph use?

Codegraph is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Codegraph use?

About 1.3k tokens (SKILL.md is roughly 5k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Codegraph?

Skills that share tags, products or a category with Codegraph: MCP Server Builder (shareAI-lab/learn-claude-code, 78k stars), Orca Run Replay (iflytek/skillhub, 5.2k stars), Yahoo Finance2 (gadicc/yahoo-finance2, 801 stars) and Workflow Schema Tuning (breaking-brake/cc-wf-studio, 5.4k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Codegraph?

aiguicai (a GitHub organization) maintains it in aiguicai/MCP-Gateway, which has 158 GitHub stars. The repository holds 5 skills in this directory. The repository was last updated on May 27, 2026.

Source: aiguicai/MCP-Gateway on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.

Install Codegraph

Facts

At a glance

What it does

When your agent uses it

Example prompts

Requirements

Workflow steps

What it can do on your machine

Tool permissions

Runs code

Network

Credentials

Context cost

Safety

SKILL.md

CodeGraph

IMPORTANT: How to Use This Tool

Prerequisites

Allowed Commands

Typical Workflow

Notes

Files

Compare with similar skills

Similar skills

More from aiguicai/MCP-Gateway

Related

Questions about Codegraph