yv1ing/z3r0/ghidra

Claude Code and agent skill by yv1ing/z3r0

Trust tier 2
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Use Ghidra headless analysis for authorized binary reverse engineering, decompilation, function/string/symbol export, and call graph analysis without a GUI.

Install

npx vskill@latest install yv1ing/z3r0/ghidra

SKILL.md

ghidra

Use Ghidra's analyzeHeadless tool through the bundled wrapper for automated reverse engineering. Import binaries, run analysis, decompile to C code, and export functions, strings, symbols, and call graphs.

Help first

Before constructing commands, run the wrapper help and use it as the source of truth:

bash
.agents/skills/ghidra/scripts/ghidra-analyze.sh --help

Resource paths

  • Wrapper script: .agents/skills/ghidra/scripts/ghidra-analyze.sh
  • Built-in export scripts: .agents/skills/ghidra/scripts/ghidra_scripts
  • Headless analyzer: /usr/local/bin/analyzeHeadless

Quick reference

TaskCommand
Full analysis with all exports.agents/skills/ghidra/scripts/ghidra-analyze.sh -s ExportAll.java -o ./output binary
Decompile to C code.agents/skills/ghidra/scripts/ghidra-analyze.sh -s ExportDecompiled.java -o ./output binary
List functions.agents/skills/ghidra/scripts/ghidra-analyze.sh -s ExportFunctions.java -o ./output binary
Extract strings.agents/skills/ghidra/scripts/ghidra-analyze.sh -s ExportStrings.java -o ./output binary
Get call graph.agents/skills/ghidra/scripts/ghidra-analyze.sh -s ExportCalls.java -o ./output binary
Export symbols.agents/skills/ghidra/scripts/ghidra-analyze.sh -s ExportSymbols.java -o ./output binary

Usage rules

bash
.agents/skills/ghidra/scripts/ghidra-analyze.sh [options] <binary>

Always call the wrapper with its .agents/skills/ghidra/... path. It handles project creation/cleanup and provides a simpler interface to analyzeHeadless.

  • Work only on provided binaries or explicitly authorized artifacts.
  • Prefer ExportAll.java for first-pass analysis unless the task needs a narrower export.
  • Use task-scoped output directories and save logs instead of streaming large decompiler output into the conversation.
  • Specify --processor and --cspec when auto-detection is wrong or firmware context is known.
  • Use --timeout for large or hostile samples and --keep-project only when the task requires later project reuse.

Options:

  • -o, --output <dir> - Output directory for results (default: current dir)
  • -s, --script <name> - Post-analysis script to run (can be repeated)
  • -a, --script-args <args> - Arguments for the last specified script
  • --script-path <path> - Additional script search path
  • -p, --processor <id> - Processor/architecture (e.g., x86:LE:32:default)
  • -c, --cspec <id> - Compiler spec (e.g., gcc, windows)
  • --no-analysis - Skip auto-analysis (faster, but less info)
  • --timeout <seconds> - Analysis timeout per file
  • --keep-project - Keep the Ghidra project after analysis
  • --project-dir <dir> - Directory for Ghidra project (default: /tmp)
  • --project-name <name> - Project name (default: auto-generated)
  • -v, --verbose - Verbose output

Built-in export scripts

ExportAll.java

Comprehensive export - runs all other exports and creates a summary. Best for initial analysis.

Output files:

  • {name}_summary.txt - Overview: architecture, memory sections, function counts
  • {name}_decompiled.c - All functions decompiled to C
  • {name}_functions.json - Function list with signatures and calls
  • {name}_strings.txt - All strings found
  • {name}_interesting.txt - Functions matching security-relevant patterns
bash
.agents/skills/ghidra/scripts/ghidra-analyze.sh -s ExportAll.java -o ./analysis firmware.bin
ExportDecompiled.java

Decompile all functions to C pseudocode.

Output: {name}_decompiled.c

bash
.agents/skills/ghidra/scripts/ghidra-analyze.sh -s ExportDecompiled.java -o ./output program.exe
ExportFunctions.java

Export function list as JSON with addresses, signatures, parameters, and call relationships.

Output: {name}_functions.json

json
{
"program": "example.exe",
"architecture": "x86",
"functions": [
{
"name": "main",
"address": "0x00401000",
"size": 256,
"signature": "int main(int argc, char **argv)",
"returnType": "int",
"callingConvention": "cdecl",
"isExternal": false,
"parameters": [{"name": "argc", "type": "int"}, ...],
"calls": ["printf", "malloc", "process_data"],
"calledBy": ["_start"]
}
]
}
ExportStrings.java

Extract all strings (ASCII, Unicode) with addresses.

Output: {name}_strings.json

bash
.agents/skills/ghidra/scripts/ghidra-analyze.sh -s ExportStrings.java -o ./output malware.exe
ExportCalls.java

Export function call graph showing caller/callee relationships.

Output: {name}_calls.json

Includes:

  • Full call graph
  • Potential entry points (functions with no callers)
  • Most frequently called functions
ExportSymbols.java

Export all symbols: imports, exports, and internal symbols.

Output: {name}_symbols.json

Common workflows

Analyze an unknown binary
bash
# Create output directory
mkdir -p ./analysis
# Run comprehensive analysis
.agents/skills/ghidra/scripts/ghidra-analyze.sh -s ExportAll.java -o ./analysis unknown_binary
# Review the summary first with bounded reads
sed -n '1,160p' ./analysis/unknown_binary_summary.txt
# Look at interesting patterns (crypto, network, dangerous functions)
sed -n '1,160p' ./analysis/unknown_binary_interesting.txt
# Check specific decompiled functions
grep -A 50 "encrypt" ./analysis/unknown_binary_decompiled.c
Analyze firmware
bash
# Specify ARM architecture for firmware
.agents/skills/ghidra/scripts/ghidra-analyze.sh \
-p "ARM:LE:32:v7" \
-s ExportAll.java \
-o ./firmware_analysis \
firmware.bin
Quick function listing
bash
# Just get function names and addresses (faster)
.agents/skills/ghidra/scripts/ghidra-analyze.sh --no-analysis -s ExportFunctions.java -o . program
# Parse with jq
jq '.functions[] | "\(.address): \(.name)"' program_functions.json
Find specific patterns
bash
# After running ExportDecompiled, search for patterns
grep -n "password\|secret\|key" output_decompiled.c
grep -n "strcpy\|sprintf\|gets" output_decompiled.c
Analyze multiple binaries
bash
for bin in ./samples/*; do
name=$(basename "$bin")
.agents/skills/ghidra/scripts/ghidra-analyze.sh -s ExportAll.java -o "./results/$name" "$bin"
done

Architecture and processor IDs

Common processor IDs for the -p option:

ArchitectureProcessor ID
x86 32-bitx86:LE:32:default
x86 64-bitx86:LE:64:default
ARM 32-bitARM:LE:32:v7
ARM 64-bitAARCH64:LE:64:v8A
MIPS 32-bitMIPS:BE:32:default or MIPS:LE:32:default
PowerPCPowerPC:BE:32:default

Find all available processors:

bash
ls /opt/ghidra/Ghidra/Processors/

Troubleshooting

Ghidra not found

Inside the sandbox, GHIDRA_HOME is /opt/ghidra and the wrapper should find /usr/local/bin/analyzeHeadless. If running the skill outside the image, set GHIDRA_HOME explicitly:

bash
export GHIDRA_HOME=/path/to/ghidra_PUBLIC
.agents/skills/ghidra/scripts/ghidra-analyze.sh ...
Analysis takes too long
bash
# Set a timeout (seconds)
.agents/skills/ghidra/scripts/ghidra-analyze.sh --timeout 300 -s ExportAll.java binary
# Skip analysis for quick export
.agents/skills/ghidra/scripts/ghidra-analyze.sh --no-analysis -s ExportSymbols.java binary
Out of memory

Set a larger Ghidra heap before running the wrapper:

bash
export MAXMEM=4G
Wrong architecture detected

Explicitly specify the processor:

bash
.agents/skills/ghidra/scripts/ghidra-analyze.sh -p "ARM:LE:32:v7" -s ExportAll.java firmware.bin

Tips

  1. Start with ExportAll.java - It gives you everything and the summary helps orient you
  2. Check the interesting.txt file - It highlights security-relevant functions automatically
  3. Use jq for JSON parsing - The JSON exports are designed to be machine-readable
  4. Decompilation isn't perfect - Use it as a guide, cross-reference with disassembly
  5. Large binaries take time - Use --timeout and consider --no-analysis for quick scans

Output

Report the artifact path, wrapper command used, output directory, key exported files, architecture or processor assumptions, relevant findings, and limitations such as timeout, decompiler failures, or unresolved processor selection.

Security

  • Security scanPassed, 98 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
573
Repository forks
97
7-day trend score
32

Stars and forks are for the whole GitHub repository, not this skill alone. Refreshed 7 Oct 2026, 22 days ago.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/yv1ing/z3r0/ghidra/badge)](https://verified-skill.com/skills/yv1ing/z3r0/ghidra)

Something wrong with this skill or its page? Tell us.