# RootSign > Developer documentation for RootSign, the open-source Python SDK that records AI agent tool calls on a tamper-evident hash chain. - [RootSign: tamper-evident audit trails for AI agents](https://docs.getprovidex.com/index.md): RootSign records every tool call your AI agent makes on a SHA-256 hash chain, so you can prove later what the agent did and that nobody edited the record. - [Quickstart: record and verify your first agent session](https://docs.getprovidex.com/quickstart.md): Install RootSign, record two traced tool calls to a local file, verify the hash chain, then edit and delete records to see TAMPERED and INCOMPLETE. - [How RootSign's hash chain detects edited and missing records](https://docs.getprovidex.com/how-it-works.md): What a RootSign action record contains, which fields the SHA-256 chain covers, what VALID, TAMPERED and INCOMPLETE mean, and what verification cannot detect. - [Record LangGraph and LangChain tool calls](https://docs.getprovidex.com/integrations/langgraph.md): Instrument the tools in a LangGraph agent with rootsign.wrap_tools() so every ToolNode call lands on a tamper-evident hash chain, with no change to your graph. - [Record CrewAI tool calls on a hash chain](https://docs.getprovidex.com/integrations/crewai.md): Instrument a CrewAI agent's tools with rootsign.wrap_crewai_tools() so every tool call in a crew run lands on a tamper-evident hash chain. - [Record MCP tool calls with the RootSign proxy and audit server](https://docs.getprovidex.com/integrations/mcp.md): Put RootSign's proxy between any MCP client and server to record every tools/call on a hash chain, and expose the audit log to agents as a read-only MCP server. - [Record calls to any Python function with @rootsign.trace](https://docs.getprovidex.com/integrations/python.md): Use the @rootsign.trace() decorator to put calls to any Python function on a RootSign hash chain, with no agent framework required. - [Verify a session's hash chain from the CLI or Python](https://docs.getprovidex.com/guides/verify.md): Check that a RootSign session is intact with rootsign verify or verify_session_local(), read the four exit codes, and run verification in CI or cron. - [Export a session as an evidence bundle for a reviewer](https://docs.getprovidex.com/guides/evidence-bundles.md): Use rootsign export to turn a session into a folder with an HTML report, JSON evidence and a SHA-256 manifest, then check that a bundle you receive is genuine. - [Require human approval before a tool runs](https://docs.getprovidex.com/guides/human-approval.md): Hold a high-risk tool call until a person approves it with rootsign approve, and record who decided, why, and what they saw, next to the action. - [Redact personal data before it is hashed and stored](https://docs.getprovidex.com/guides/redaction.md): Use StandardPIIConfig and the other redaction configs to replace emails, phone numbers and other PII with [REDACTED] in tool inputs and outputs before RootSign hashes them. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.