# Facticity AI

Real-time, multi-modal fact-checking for AI systems.

## Quick Start

```bash
# Connect this server (installs CLI if needed)
npx -y @smithery/cli@latest mcp add ai-seer/facticityai

# Browse available tools
npx -y @smithery/cli@latest tool list ai-seer/facticityai

# Get full schema for a tool
npx -y @smithery/cli@latest tool get ai-seer/facticityai fact_check

# Call a tool
npx -y @smithery/cli@latest tool call ai-seer/facticityai fact_check '{}'
```

## Direct MCP Connection

Endpoint: `https://facticityai--ai-seer.run.tools`

## Tools (8)

- `fact_check` — Fact-check a claim. Returns classification (True/False), assessment, evidence, and sources.
- `extract_claim` — Extract claims from text or video URLs (YouTube, TikTok, Instagram). Auto-transcribes video content.
- `get_credits` — Check remaining API credits and account information
- `check_task_status` — Check the status of an async fact-check task using its task_id
- `get_more_credits` — Explains how to buy/restore credits. Use this when you run out of credits or get billing/credit errors.
- `how_to_use` — Explains what Facticity.AI is, its strengths, and common use-cases. Good first tool for onboarding.
- `transcribe_link` — Transcribe content from a URL (YouTube, TikTok, Instagram videos)
- `link_reliability_check` — Check the reliability and bias of a URL using MediaBias data. Returns bias score, quality score, and labels.

```bash
# Get full input/output schema for a tool
npx -y @smithery/cli@latest tool get ai-seer/facticityai <tool-name>
```

## Resources

- `resource://facticity/homepage` — Overview and dashboard
- `resource://facticity/api_docs` — Endpoints and usage for fact_check, extract_claim, and more

## Prompts (2)

- `fact_check_best_practices` — Guidance for writing high-quality fact-check queries
- `onboarding` — Short intro and recommended first steps