What is clangd-graph-rag?

A semantic and structural search tool for C and C++ codebases backed by a graph database. It maps relationships such as function calls, class inheritance, macros, and type aliases.

Use it to explore project architecture, trace caller and callee chains, inspect C++ inheritance and templates, follow aliases and macros, and search for code by meaning. It requires a prepared Neo4j graph and configured server.

It helps answer questions that ordinary text search can miss, such as where a function is called, how a class is inherited, or what a macro-generated symbol refers to.

clangd-graph-rag — Skill for Claude Code and Codex by 2015xli

Install

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.

Any agent

npx skills add 2015xli/clangd-graph-rag --skill clangd-graph-rag

Clone the repo

git clone --depth 1 https://github.com/2015xli/clangd-graph-rag

Made for: Claude Code, Codex.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/2015xli/clangd-graph-rag/clangd-graph-rag/github.svg)](https://agentmods.dev/skills/2015xli/clangd-graph-rag/clangd-graph-rag)

Your own site

<a href="https://agentmods.dev/skills/2015xli/clangd-graph-rag/clangd-graph-rag"><img src="https://agentmods.dev/badge/skills/2015xli/clangd-graph-rag/clangd-graph-rag/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/2015xli/clangd-graph-rag/clangd-graph-rag"><img src="https://agentmods.dev/badge/skills/2015xli/clangd-graph-rag/clangd-graph-rag.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 0 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 834 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00000	$0.00834
Opus 5	$0.00000	$0.00417
Sonnet 5	$0.00000	$0.00167
Haiku 4.5	$0.00000	$0.00083

Measured 11d ago against content hash bf0f031cb431, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

clangd-graph-rag 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 11d 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.

SKILL.md · 48 lines

How it starts

The opening of the file, as written. The whole thing — 48 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Skill: clangd-graph-rag

This skill enables deep semantic and structural analysis of C/C++ codebases using a pre-built Neo4j GraphRAG. It provides insights into call chains, class hierarchies, macro causality, and type aliases.

Activation

Activate this skill when the user asks questions about:

Project architecture, module responsibilities, or high-level workflows.
Call chains (caller/callee relationships) or method overriding.
C++ inheritance structures and template specializations.
"Magic symbols" generated by macros or complex typedef/using alias chains.
Semantic code search (e.g., "Find the logic for packet validation").

Setup Requirements

A Neo4j database populated by the clangd-graph-rag pipeline.
The graph_mcp_server.py must be configured as an MCP server.
Environment variables: NEO4J_URI, NEO4J_USER, NEO4J_PASSWORD.

Core Instructions

1. Orientation & Discovery

Always start by calling get_project_info and get_graph_schema.
get_project_info provides the path (absolute project root) and a high-level summary.
get_graph_schema explains the node labels (e.g., FUNCTION, CLASS_STRUCTURE, MACRO, TYPE_ALIAS) and relationships.
Paths: All path properties in the graph are relative to the project root.

2. Structural Querying (Cypher)

Use execute_cypher_query for precise structural analysis.
Semantic Labels: Prefer specific labels (FUNCTION, METHOD, CLASS_STRUCTURE, MACRO, TYPE_ALIAS) over the generic ENTITY label for efficiency.
Macros: Follow (s)-[:EXPANDED_FROM]->(m:MACRO) to explain symbols generated by the preprocessor. Check the original_name property on the symbol for the raw invocation text.
Types: Follow (ta:TYPE_ALIAS)-[:ALIAS_OF]->(t) to resolve alias chains (e.g., MyInt2 -> MyInt -> int).
Calls: Use SHORTEST path selectors (e.g., MATCH p = SHORTEST 5 (a:FUNCTION)-[:CALLS*]->(b:FUNCTION)) to prevent result set explosion.
Result Management: Always use LIMIT (e.g., LIMIT 10) on custom queries.

Read the full file on GitHub · 48 lines

Changes

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.

11d ago First seen · 48 lines · 0 tokens per session scan A bf0f031cb431

Subscribe to this mod's changes

clangd-graph-rag is a skill published in the GitHub repository 2015xli/clangd-graph-rag (63 stars, last pushed 4mo ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 834 tokens. 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 2026-08-30.

C and C++Data and AI Languages RAG and vectors clang clangd code-graph knowlege-graph llm-rag source-code-analysis

ax-cpp-ai

Use when writing C++ code with axllm for named deployment profiles, generic provider clients, model selection, OpenAI-compatible calls, Responses, Gemini, Anthropic, routers, and balancers.

ax-llm/ax · 47 tokens

ax-cpp-gen

Use when writing C++ code with axllm for AxGen programs, forward calls, indexed multi-sampling, result pickers, streaming, tools, assertions, traces, usage, and output parsing.

ax-llm/ax · 48 tokens

ax-cpp-llm

Use when writing C++ code with axllm for using the generated Ax package, factory functions, package docs, examples, and API reference.

ax-llm/ax · 38 tokens

ax-cpp-signature

Use when writing C++ code with axllm for string signatures, field descriptors, JSON schema output, validation, and typed tool argument shapes.

ax-llm/ax · 37 tokens

ax-cpp-gepa

Use when writing C++ code with axllm for GEPA, Pareto tradeoffs, reflection clients, metric budgets, optimizer state, and artifacts.

ax-llm/ax · 39 tokens

cuda-kernels

Provides guidance for writing and benchmarking optimized CUDA kernels for NVIDIA GPUs (H100, A100, T4) targeting HuggingFace diffusers and transformers libraries. Kernels must be kernel-builder/ABI3-compliant: no pybind11, no setup.py, TORCHLIBRARYEXPAND bindings only. Supports models like LTX-Video, Stable Diffusion…

huggingface/kernels · 120 tokens