Mnestiq

Mnestiq is a flight recorder for AI agents that captures signed, tamper-evident evidence of agent actions.

Mnestiq is a flight recorder designed for AI agents to capture tamper-evident and signed evidence of what an agent saw, decided, and did. It helps with incident response by recording model calls, tool calls, and network connections. The tool tracks attribution, egress identity, provenance of model context, and hash-chains records with Ed25519-signed Merkle checkpoints. It wraps Anthropic and OpenAI clients via SDK capture or records from outside the agent using a proxy. The local investigation dashboard features a step-by-step timeline, provenance-colored context, verification status, and a connection finder.

Mnestiq at a glance

Best for: Software engineer
Runs on: API
Pricing: Free
Released: October 2026
Category: AI agent tool
Website: github.com/mnestiq/mnestiq

Key features

Attribution tracking for runs, steps, agents, and operators
Egress identity recording for outbound HTTP and raw TCP connections
Model context provenance tagging
Tamper-evident hash-chaining with signed Merkle checkpoints
Evidence redaction preserving verification
Drop-in SDK capture for Anthropic and OpenAI clients
Proxy recording from outside the agent

Use it for

Investigating agent security incidents and prompt injections
Tracing outbound network connections to agent steps
Verifying tamper-evident records of agent execution
Sharing redacted evidence with auditors

Alternatives to Mnestiq

flightdeck — Monitor and control production and coding agents with real-time token budgets and kill switches.
ARTEX — Manages autonomous penetration testing tasks and tracks vulnerability progress.
Aegis Kernel — Stops prompt injection and tool-hijacking at the OS syscall boundary.
AgentScore — Monitors production agents and dispatches coding agents to fix failures automatically.
Playground — Persists agent memory across sessions and searches by meaning.
Composio — Equips AI agents with tools, managed user auth, and sandboxes across 1,500+ apps
Supermemory — Integrates long-term memory into AI coding agents with a single prompt.
Memvid — Single-file architecture replaces complex RAG pipelines with portable storage.
FakeLiang — Verifies attack paths step by step with traceable call records and evidence.
Melra — Governs AI agent tool calls with local policies, approval gates, and verified receipts.
APIStack — Connect AI agents to over 100 tools using one API key or a single MCP endpoint.
Agent Beacon — Retains coding agent knowledge across tools so errors do not repeat.

All Mnestiq alternatives