debug-cpp

A C and C++ debugging workflow using LLDB, a debugger that lets you pause a program and inspect its execution.

In plain words
What is it for?
Use it to build programs with debugging information, set breakpoints, step through code, inspect variables, and view call stacks.
Why use it?
It helps investigate crashes, incorrect values, hangs, and unexpected output by examining the program while it runs.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/benpm/claude_lldb_mcp/debug-cpp
Any agent
npx skills add benpm/claude_lldb_mcp --skill debug-cpp
Clone the repo
git clone --depth 1 https://github.com/benpm/claude_lldb_mcp

Made for: Claude Code, Codex.

Per session 173 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,060 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00173 $0.02060
Opus 5 $0.00086 $0.01030
Sonnet 5 $0.00035 $0.00412
Haiku 4.5 $0.00017 $0.00206

Measured yesterday against content hash 5bfcf1c1d050, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

debug-cpp 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 yesterday.

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.

skills/debug-cpp/SKILL.md · 130 lines

How it starts

The opening of the file, as written. The whole thing — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Debugging C/C++ with LLDB MCP Tools

This skill guides you through debugging compiled C/C++ programs using the LLDB MCP server. The server provides 17 specialized tools that wrap LLDB in batch mode — each tool call spawns a fresh LLDB process, so there's no persistent session state between calls. This means you need to replay setup (target + breakpoints) on each tool invocation, which the tools handle automatically via their parameters.

Step 0: Gather what you need

Before touching any tool, ask the user (or infer from context) these three things:

  • Executable path — the compiled binary (absolute path is most reliable)
  • What's wrong — crash, wrong output, hang, assertion failure, unexpected variable value?
  • Arguments — command-line args the program needs (can be empty)

If the user only has source files, go to Build first below. If they already have a binary, skip ahead.

Build first (if no binary)

Compile with debug symbols and no optimization — without -g -O0, LLDB can't show source lines or variable names:

clang++ -g -O0 -o /abs/path/to/output /abs/path/to/source.cpp
# Multi-file:
clang++ -g -O0 -o output main.cpp helper.cpp
# CMake project:
cmake -B build -DCMAKE_BUILD_TYPE=Debug -DCMAKE_EXPORT_COMPILE_COMMANDS=ON
cmake --build build

Windows/git bash path note: LLDB on Windows accepts both Unix-style git bash paths (/c/Users/ben/proj/app) and Windows-style paths (C:\Users\ben\proj\app). Pass whichever form the user gives you — both work. Use absolute paths to avoid working-directory ambiguity.

Step 1: Confirm the environment

Call lldb_version (no parameters) to confirm LLDB is installed and reachable. If this fails, LLDB is not in PATH — on Windows, the LLVM installer must add C:\Program Files\LLVM\bin to the system PATH.

Step 2: Choose your entry point

Decide where to stop execution first based on what you're investigating:

Scenario Breakpoint approach
Function crashes or has wrong output location = "function_name"
Bug at a specific source line location = "/abs/path/file.cpp:42"
Program crashes immediately Use lldb_analyze_crash instead of breakpoints
Want to start from scratch location = "main"
Address you found in a crash log location = "0x401234"

Read the full file on GitHub · 130 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.

  1. yesterday First seen · 130 lines · 173 tokens per session scan A 5bfcf1c1d050

Subscribe to this mod's changes

debug-cpp is a skill published in the GitHub repository benpm/claude_lldb_mcp (3 stars, last pushed 5mo ago), licensed Apache-2.0. It adds 173 tokens to every session and 2,060 once invoked, about $0.0009 per session on Opus 5. 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 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens