Evidence Python Packages

nkama-fact-benchmark

Evidence-gated benchmark for testing whether AI assistants can prove what they claim

keel-verifier

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

sclite-core

Schema-backed contract layer for governed AI/security workflows: intent, policy decisions, approvals, execution receipts, evidence, and replayable audit traces.

qs-dmss

QS-DMSS Studio: evidence-first simulation lab for reproducible runs, evidence bundles, campaign comparison, and research-object exports.

falsiflow

Stop unverifiable AI eval, product metric, and R&D claims from passing CI.

assay-ai

Signed, offline-verifiable evidence for tool-using AI systems.

ravenclaw-security

Governance-first security research runtime with policy-gated execution and public-safe proof artifacts

motusos

Local event-backed work ledger and Store/run/event/projection reference CLI.

repro-evidence-kit

Reproducible artifact manifests, sandbox-output verification, and evidence-bundle validation for artifact-heavy workflows.

research-md

The decision forge — evidence-graded, phase-gated, peer-reviewed decisions

delivery-workbench

Evidence-first delivery framework for agentic software work: roadmaps, commit contracts, work logs, and adoption discovery.

tfs-test-runner

Convert Azure DevOps / TFS xlsx test case exports into a self-contained HTML execution kit with screenshot capture, status tracking, and PDF evidence export. Optional GPT translation.

emet

A small external witness for AI source and view consistency: re-derive the bytes and get MATCH, DRIFT, or UNVERIFIABLE, never TRUSTED. Four independent implementations (Python, Rust, Node, Go) pass the same frozen conformance vectors. Zero dependencies.

cortexpilot-orchestrator

OpenVibeCoding is the command tower for AI engineering: plan, delegate, track, resume, and prove long-running work across Codex and Claude Code.

evidential-deep-learning

Learn fast, scalable, and calibrated measures of uncertainty using neural networks!

openvibecoding-orchestrator

OpenVibeCoding is the command tower for AI engineering: plan, delegate, track, resume, and prove long-running work across Codex and Claude Code.

agent-cooking-cli

Evidence-first, artifact-driven bug triage workflow for CLI agents

cra-sbom-evidence

CRA Article 14 evidence pack from your SBOM and VEX feeds, verbatim-cited and hash-chained.

langgraph-oep

Capture decision-id-joined evidence from LangGraph agents. Wraps your checkpoint saver to emit OEP tool_permission_packet.v0 records covering the 19 surfaces LangGraph's StateSnapshot does not bind natively.