# Logic-LM (Answer Set Programming)

> Use this tool when you need to perform formal logical reasoning, solve constraint satisfaction problems, or tackle complex multi-step logical deductions that require symbolic reasoning capabilities. It translates natural language problems into Answer Set Programming (ASP) code, executes reasoning with the Clingo solver, and interprets results back to natural language. Ideal for applications such as academic research, automated theorem proving, and tasks where traditional language model reasoning is insufficient.

Canonical page: https://skillsregistry.net/skills/logic-lm  
JSON: https://api.skillsregistry.net/v1/skills/logic-lm

## Description

Logic-LM MCP server by Steven Wang that provides AI assistants with symbolic reasoning capabilities through Answer Set Programming (ASP) and the Clingo solver. Built with Python using FastMCP, the implementation follows a three-stage Logic-LM pipeline: translating natural language problems to ASP code, executing symbolic reasoning with Clingo, and interpreting results back to natural language. The server includes comprehensive ASP translation guidelines, template libraries for common logical patterns (syllogisms, conditionals, universal quantification), and tools for direct ASP program verification with self-refinement capabilities. Designed for applications requiring formal logical reasoning, constraint satisfaction problems, and symbolic problem-solving where traditional LLM reasoning may be insufficient, particularly useful for academic research, automated theorem proving, and complex multi-step logical deduction tasks.

## Trust

- **Trust score (0–1):** 0.50
- **Verification tier:** unverified

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** devops-ci
- **Updated:** 2026-04-29

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/logic-lm)
- **Repository:** <https://github.com/shipitsteven/logic-lm-mcp>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "logic-lm"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/logic-lm` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/logic-lm/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt