Agent Reliability Python Packages

superlocalmemory

Open-source governed, local-first memory control plane for AI agents and teams. arXiv:2608.08253

agentraas

The reliability layer for what AI agents do: exactly-once execution, action policies, human approval and a full audit trail. Self-hosted, MCP-native, free Community edition (MIT/Apache-2.0).

house-party-protocol

A local harness for agent teams on Claude Code and Codex CLI: only what they prove counts as done. Waves and lanes split the work, reviewers from another model family judge it, councils keep the dissent that lost, typed decisions with Jev are measured before they are trusted, and approvals are bound to the bytes they approved. No model calls.

nabit

nab your agent's silent failures — verify LLM-agent side effects against real system state, with self-heal. Zero deps.

fathom-read

Catch the step where an AI agent contradicts its own earlier decision. CLI and adapters for LangGraph, CrewAI, Letta, OpenInference, DBOS and multi-agent logs, plus an expiry read for long-running agents.

sourcerykit

Python SDK for accurate and verifiable agent tool use. Agents verify that answers came from the right source and were not changed. Downstream agents detect 100% of errors and retry to achieve a 50% jump in answer accuracy.

failproofai-sdk

Observability and enforcement for AI agent harnesses. Capture every run and runtime reliability with policy enforcement.

kalibr

Stop overpaying to run your agents. Kalibr routes every request to lower-cost model and tool paths without degrading performance.

freshctx

Pre-action evidence revalidation for AI agents. Detect stale reasoning before agents write, send, update, or delete.

maida-ai

Pre-merge behavioral regression gate for AI agents. Check agent changes, inspect regressions, and block broken behavior before merge. Local-first.

loopbrake

Brakes for stuck AI agents. Learns how long your agent's tasks normally take, stops the one that runs way past that, and says why. For Claude Code, Codex and your own Python agents. Runs locally, zero dependencies.

dunetrace

The runtime reliability layer for AI agents.

langchain-fathom

LangChain agent middleware that flags the step where an agent contradicts its own earlier action. Deterministic, no model calls, fits tests and CI.

agent-signage

Road signs for coding agents: one measured fact at the moment of action, silence otherwise

agentassert-abc

Runtime behavioral contracts for AI agents: define YAML contracts, check adapter outputs, and raise explicit contract violations. Agent Behavioral Contracts (ABC).

fp-cloud-cli

Observability and enforcement for AI agent harnesses. Capture every run and runtime reliability with policy enforcement.

polycog-cognition

Reasoning foundation for trustworthy agents.

agentdbg

Pre-merge behavioral regression gate for AI agents. Check agent changes, inspect regressions, and block broken behavior before merge. Local-first.

langchain-right-rudder

LangChain agent middleware that flags the step where an agent contradicts its own earlier action. Deterministic, no model calls, fits tests and CI.

agent-claimcheck

Catch AI agents that say "done" when the task isn't. Checks agent-trace/v1 traces with claim-evidence rules, LLM judges and a trained classifier, returns verified / false_success / unverifiable with calibrated confidence, and ships a 300-trace labelled benchmark.

pydantic-ai-trustabl

Trustabl turns Pydantic AI apps into governed agents. It verifies typed tools, structured outputs, usage limits, and idempotent mutations, repairs production reliability gaps, generates least-privilege policies, and adds OTEL context when a run still slips the schema.

right-rudder

Catch the step where an AI agent contradicts its own earlier decision. CLI and adapters for LangGraph, CrewAI, Letta, OpenInference, DBOS and multi-agent logs, plus an expiry read for long-running agents.

self-heal-llm

Automatic repair for failing Python code, powered by any LLM.

auditable

Audit any agent decision across its past, present, and future, on one typed graph.