What Your Agent Can Do
Structured, evidence-backed results through a simple API — not raw files the agent can't process.
Full-text search across all documents with faceted filtering by entity type, date, file type, and more.
Access 30+ automatically extracted entity types — people, organizations, emails, phones, crypto addresses, bank accounts, dates.
Find connection paths between any two entities across the corpus. Retrieve the specific documents where two entities co-occur.
Retrieve full document content, structured data (CSV, Parquet, JSON as clean arrays), and file metadata.
Upload new document sets through an automated ETL pipeline. Executable Python scripts handle complex workflows like bulk ingestion.
Real-time pipeline status with per-phase progress and per-endpoint readiness indicators.
What Makes It Different
Self-Teaching
The agent reads the full developer guide from the API before operating. No prompt engineering, no hardcoded instructions. The platform teaches the agent how to use it.
No other tool in agent registries offers this.
Processing-Aware Responses
Every API response includes corpus readiness signals. When data is still being processed, the agent knows — and tells the user instead of hallucinating conclusions from incomplete data.
Anti-hallucination at the infrastructure level.
Delegated User Identity
The agent inherits the human user's full identity — organization membership, investigation access, policy evaluation. Every action is traceable to a specific authenticated user.
Enterprise audit trail built in.
Available on ClawHub
Two skills on ClawHub, the public skill registry for OpenClaw agents. Same platform, two positioning lenses.
Investigative intelligence for AI agents
For investigators, compliance teams, forensic accountants, and anyone building agents that need to search evidence, extract entities, and trace relationships across mixed-format documents.
- Search, extract, connect, and verify across evidence
- 30+ entity types with relationship graphing
- Self-teaching API + executable scripts
Document ETL and entity relationship engine for AI agents
For agent builders, knowledge automation teams, ops/research teams, and anyone who needs to give an AI agent access to a document corpus without burning tokens on raw parsing.
- 1,000+ file formats with structured data support
- Entity extraction + relationship graphing
- Processing-aware, anti-hallucination guardrails
Get Started in Minutes
Create an Account
Register and verify your email. Set up MFA with any authenticator app.
Generate Credentials
Visit the agentic token page. Click Generate Credentials. Copy the two environment variables.
Connect Your Agent
Set INGESTIGATE_TOKEN and INGESTIGATE_BASE_URL in your agent's platform settings.
Security Model
Built for industries where data security isn't optional.
- Short-lived access tokens (30-minute expiry)
- No persistent API keys — by design
- MFA required on all accounts
- Organization-scoped data isolation
- Delegated user identity — full audit trail
- Air-gapped deployment available
Agentic API Included at No Extra Cost
The agentic API is included with every Ingestigate plan. No separate pricing, no add-on fees.
View PlansReady to Connect Your Agent?
Free trial. No credit card. Your agent is searching documents in minutes.