skills.rest
doodledood
doodledoodCOMMUNITY
Explore all skills by this author →
GitHub

consultant

Run asynchronous code analyses with LiteLLM-powered AI models.

17|4|Updated Nov 20, 2025
Search Tags:#code analysis#security audit#LLM#Python#architectural review#asynchronous#LiteLLM
One-click install
npx skills add https://github.com/doodledood/claude-code-plugins --skill consultant
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: consultant
Source: https://github.com/doodledood/claude-code-plugins/tree/main/claude-plugins/consultant/skills/consultant
Command: npx skills add https://github.com/doodledood/claude-code-plugins --skill consultant

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires litellm, requests, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill provides expert-level AI analysis for complex development tasks, automating architectural reviews, security audits, and comprehensive code understanding across large codebases. It saves developers significant time and reduces the complexity of gaining deep insights that might otherwise be missed or take extensive manual effort.

Core Features & Use Cases

  • Multi-Provider LLM Support: Access over 100 AI models via LiteLLM, including OpenAI, Anthropic, Google, and local models, with custom base URLs.
  • Asynchronous Operation: Start complex analyses in the background and reattach later, allowing you to continue other work without interruption.
  • Token & Context Management: Automatically counts tokens, validates against model context limits, and provides clear suggestions if limits are exceeded.
  • Automatic Model Selection: Intelligently discovers and selects the best available model based on capabilities and context size.
  • Use Case: Initiate a background security audit on all Python files in your src directory, receiving a detailed report on vulnerabilities and recommended fixes without blocking your terminal.

Quick Start

Start a background consultation to analyze Python files for security vulnerabilities: python3 claude-plugins/consultant/skills/consultant/scripts/oracle_cli.py
--prompt "Analyze this code for security vulnerabilities"
--file src/**/*.py
--slug "security-audit"

Frequently Asked Questions about consultant

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

FAQPage Schema
How do I analyze Python code for security vulnerabilities across large codebases?▼

Use asynchronous LLM-powered analysis to scan multiple Python files for security issues. The Skill automates vulnerability detection across your codebase through 100+ AI models via LiteLLM, returning detailed findings and recommended fixes without blocking your workflow.

Can I run code analysis in the background and check results later?▼

Yes. The Skill supports asynchronous operation—start a background consultation with a session slug, then reattach later to retrieve results. This lets you initiate complex architectural reviews or security audits and continue other work without waiting.

How do I ensure my code analysis stays within model context limits?▼

The Skill automatically counts tokens and validates against each model's context ceiling. If your code exceeds limits, it provides clear suggestions for splitting analysis or selecting a model with larger context windows.

What AI models and providers does this code analysis support?▼

The Skill accesses 100+ models across OpenAI, Anthropic, Google, and other providers via LiteLLM. You can explicitly select a model, enable automatic model selection based on capabilities, or specify custom base URLs for local or proprietary models.

Can I use this for architectural decisions and code review at scale?▼

Yes. The Skill handles comprehensive code reviews, architectural assessments, and pattern analysis across large Python codebases. Asynchronous execution and robust file handling let you process extensive repositories and receive expert-level insights without manual effort.

Do I need specific dependencies installed to run this Skill?▼

The Skill requires LiteLLM for multi-provider LLM access and requests for HTTP operations. Both are included as dependencies and handle model routing, authentication, and session management automatically.

Related Skills

Explore More Software Engineering Skills→
florianbuetowflorianbuetow
12

report

Generate security reports from scanner findings in Markdown, HTML, JSON, and SARIF formats.

Community
Advanced
peancharoenpeancharoen

speckit.checker

Run static analysis tools to detect linting, type, and security issues across source code.

Community
Advanced
0xdeafcafe0xdeafcafe

review-hygiene

Audit changed code with check-only linting and formatting tools.

Community
Advanced
Devsoul2026Devsoul2026
150

openhands

Delegates coding tasks to the OpenHands CLI via LiteLLM providers, outputting JSON-formatted results.

Community
Intermediate