cli-anything-lldb

Automate stateful LLDB debugging workflows and output JSON results.

Updated May 7, 2026
One-click install
npx skills add https://github.com/dawsonblock/JUDGE_ATLAS --skill cli-anything-lldb-dawsonblock
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cli-anything-lldb
Source: https://github.com/dawsonblock/JUDGE_ATLAS/tree/main/external/CLI-Anything-main/lldb/agent-harness/cli_anything/lldb/skills
Command: npx skills add https://github.com/dawsonblock/JUDGE_ATLAS --skill cli-anything-lldb-dawsonblock

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It solves the problem of performing repeatable, multi-step native debugging in LLDB without losing state, context, or machine-readable results.

Core Features & Use Cases

  • Stateful LLDB workflows: Run debug target creation, process launch/attach, and interactive sessions while preserving session state across steps.
  • Breakpoint and execution control: Set, list, delete, enable, disable breakpoints; continue execution; and perform step operations (over/into/out).
  • Inspection and analysis with JSON output: Retrieve thread lists, backtraces, frame locals, evaluate expressions, scan memory, and load core dumps for postmortem analysis.
  • Use Case: When debugging a crashing service, you can attach to a running process, set a function breakpoint, inspect locals at the fault point, and capture backtraces and memory findings in structured JSON for later analysis.

Quick Start

Run cli-anything-lldb with JSON output to create a target from an executable, launch it, and immediately fetch a backtrace for triage.

Frequently Asked Questions about cli-anything-lldb

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate LLDB debugging sessions and get JSON output?

You can automate LLDB debugging by using the LLDB Python API with click-based CLI parsing to produce machine-readable JSON results. This maintains persistent session state across commands for repeatable native program analysis.

Can I load a core dump in LLDB and extract a backtrace as structured data?

Yes, you can load core dumps for postmortem analysis and capture backtraces as structured JSON. The session evaluates expressions and retrieves frame locals while preserving the debugging context.

How do I manage breakpoints and step through a process in an automated LLDB workflow?

Automated LLDB workflows support creating targets, launching or attaching to processes, and managing breakpoints. You can set, enable, disable, or delete breakpoints and execute step operations while the state persists.

Do I need the LLDB Python API to run stateful debugging workflows?

Yes, a functioning LLDB Python API with click-based CLI parsing is required. This foundation enables stateful debugging sessions, allowing you to execute multi-step workflows and retrieve machine-readable results.

What is the best way to inspect memory and thread state during native debugging?

The best way to inspect native debugging state is to use structured LLDB commands that retrieve thread lists, scan memory, and evaluate expressions. This approach outputs findings in JSON for later analysis.

Why does my multi-step LLDB debugging workflow lose context between commands?

Standard LLDB sessions often lose context without state management. Using a stateful LLDB workflow framework preserves session state across target creation, process execution, and memory inspection steps.