Tamper Evident Python Packages

bernstein

The open‑source AI Agents Governance & Orchestration framework: write the rules declaratively, Bernstein enforces them and produces the verifiable, replayable record. Free, Apache-2.0. https://bernstein.run

inspeximus

The self-correcting memory layer for AI agents. Zero-dependency Python memory and MCP server. Supersede, revert, or forget a value deterministically. Verifiable erasure, witness-backed tamper-evident receipts, EU AI Act ready.

agora-mnemo

The self-correcting memory layer for AI agents. Zero-dependency Python memory and MCP server. Supersede, revert, or forget a value deterministically. Verifiable erasure, witness-backed tamper-evident receipts, EU AI Act ready.

tamga-protocol

Encrypted, portable, tamper-evident agent state — identity + memory + verifiable work receipts that survive the host. WASI 0.3, default-deny, simnet. PyPI: tamga-protocol

keel-verifier

Independent verifier for signed AI audit exports and Permit-spec evidence

nomosguard

Deterministic security reasoning core: hash-chained evidence ledger, attack-path rule engine, fail-closed policy gate. LLMs produce evidence and explain decisions; they never cross the decision boundary.

noirebox

The flight data recorder for AI agents — hash-chained, Ed25519-signed journal with RFC 3161 anchoring and third-party-verifiable exports. One TSA seal covers a whole fleet. MIT

halo-record

Tamper-evident audit trails for AI agents: hash-chained Runtime Records, dependency-free, verifiable by anyone.

epi-recorder

An audit trail for AI agents: portable, signed, offline-verifiable .epi evidence files with declared capture scope.

qp-capsule

The open protocol for cryptographically sealed audit records of autonomous AI operations. SHA3-256 + Ed25519 + hash chain. Spec, conformance suite, and reference implementations in Python and TypeScript.

plan-auditor

Make an AI agent's 'done' mean 'here is the command output that proves it.' Deterministic, LLM-free verification for AI coding agents: sealed plans, SHA-256 hash-chained evidence, STRIPS/PDDL reachability, and a fail-closed completion gate.

memdebug

See what your AI agent's memory holds, what changed, and put it back. Local-first and agent-neutral. Alpha.

homestead-memory

See what your AI agent actually did. A tamper-evident, hash-chained record of every tool call, in a local file. No server, no deployment. Verify it with zero installs.

provenrail

Guardrails for AI agents, with a receipt. Blocks destructive tool calls before they run and signs every decision into a hash-chained record anyone can verify with an open-source tool.

obstat

An auditable decision record for agent tool calls. The clearance is written down before the call runs.

arkhive-mcp

ArkHive — local-first governed memory for AI agents: tamper-evident SHA-256 ledger for identity, audit history, verification, and constraint-aware decisions. Apache-2.0.

sealstack

Signed, hash-chained action records for AI agents, receipts counter-signed by the receiving service, an offline verifier (exit 0/1/2), exports to AERF, Agent Receipts and noa. Python SDK + receipt spec.

tamper-signal

Tamper Signal: signed receipts for vibe-coded data pipelines. Proves nobody changed your data, and shows the exact link if they did.

tracekit-ai

Tamper-evident tracing, policy gates and cross-checks for AI agents: see what the agent did, and why.

ai-audit-trail

Cryptographically verifiable audit trail for AI systems — Ed25519 signed, hash-chained Decision Receipts with ISO 42001 / NIST AI RMF compliance. No blockchain, no SaaS.

logsiegel

Tamper-evident, privacy-preserving event logs for AI systems. Transparency-log style (RFC 6962), Ed25519 checkpoints, offline receipts, crypto-shredding. No blockchain.

deponent

A governed sovereign agent kernel — make any local AI agent testify. Deny-by-default gate, macOS Seatbelt jail, tamper-evident hash-chained ledger, verifiable receipts. Apache-2.0, zero deps.

decern

Deterministic authorization + tamper-evident audit for AI agents, humans, and workloads — one principal type, safety invariants machine-checked by an SMT solver (cvc5), every decision independently verifiable offline. Pure Rust.

gurdy

Runtime governance for AI agents: intercept agent-to-tool traffic, decide against a versioned Cedar policy bundle, and write a tamper-evident decision ledger a third party can verify offline.