parseltongue-claude
MCP server for hallucination detection in Claude Code, requiring every claim to cite verbatim quotes and flagging misquotes.
README
parseltongue-claude
MCP server + skill for Parseltongue hallucination detection in Claude Code.
Quick Start
# 1. Install
uv pip install -e .
# 2. Add MCP server
claude mcp add --transport stdio parseltongue -- $(pwd)/run-server.sh
# 3. Install skills
mkdir -p ~/.claude/skills
ln -s $(pwd)/skills/parseltongue ~/.claude/skills/parseltongue
ln -s $(pwd)/skills/parseltongue-review ~/.claude/skills/parseltongue-review
# 4. Restart Claude Code, then:
/parseltongue [document] What are the key claims?
/parseltongue-review README.md # audit your own docs
What It Does
Inverts Parseltongue's architecture: instead of Parseltongue calling an LLM API, Claude Code becomes the LLM and calls Parseltongue for verification.
Original: You → Parseltongue → LLM API → DSL → Verify
This: You → Claude Code → Parseltongue MCP → Verify
Every claim must cite a verbatim quote. Misquotes are flagged as hallucinations.
Skills
/parseltongue — Grounded document analysis
Analyze external documents with provable claims. Every fact traces to a verbatim quote.
/parseltongue [SEC filing] Are there revenue recognition red flags?
/parseltongue-review — Documentation audit
Audit docs for accuracy, duplication, and consistency with source.
/parseltongue-review README.md # vs upstream, find duplicated content
/parseltongue-review docs/ # audit mkdocs site
MCP Tools
| Tool | Description |
|---|---|
parseltongue_create_session |
Create verification session |
parseltongue_register_document |
Register source document |
parseltongue_load_dsl |
Load DSL and verify quotes |
parseltongue_check_consistency |
Full consistency check |
parseltongue_dsl_reference |
Get DSL docs |
See upstream Parseltongue docs for DSL syntax and examples.
Development
uv pip install -e ".[dev]" # install
uv run pre-commit install --hook-type commit-msg --hook-type pre-commit
uv run pytest # test (80% coverage)
uv run ruff check --fix . && uv run ruff format . # lint + format
uv run cz commit # conventional commit
License
Apache-2.0
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.