MCP-RISK-LINTER
№ 01
ARTIFACT
pip install mcp-risk-linter
Risk lint for MCP manifests.
Risk taxonomy, CLI, and GitHub Action wrapper for MCP manifests, permissions, claims, and unsafe tool surfaces.
LICENSE
MIT · public
AURAONE OPEN · V2 · TWELVE PACKAGES · MIT
Open v2 hub.
Twelve installable packages for MCP and A2A review, trace-to-regression workflows, LeRobot dataset gates, VLA robustness diagnostics, recovery metadata, and lab review packets.
READING · open · v2 · published
AGENT TRUST
MCP and A2A
Risk linting, contract tests, trace replay, OTEL bridges, trace cards.
ROBOTICS DATA
LeRobot first
Quality gates, recovery metrics, VLA robustness probes, embodiment cards.
REVIEW SURFACE
Synthetic, public
Failure cases with reproducible commands. Outreach packets without endorsement.
OPEN V2 CATALOG
Twelve packages for technical review.
Each scoped to a command-line diagnostic, portable artifact, GitHub workflow, or review packet a lab engineer can inspect without a sales call.
AGENT TRUST · 6 packages
MCP risk linting, A2A contract tests, deterministic replay, OTEL bridges, trace cards, and prompt-rubric drift review.
OPEN SOURCE →
A2A-CONTRACT-TEST
№ 02
ARTIFACT
pip install a2a-contract-test
Contract tests for A2A cards.
Offline contract tests for A2A agent cards, task lifecycle states, structured payloads, and errors.
LICENSE
MIT · public
TOOL-CALL-REPLAY
№ 03
ARTIFACT
pip install tool-call-replay
Deterministic agent replay.
Deterministic replay harness that turns failed agent tool-call traces into local regression tests.
LICENSE
MIT · public
AGENT-TRACE-CARD
№ 04
ARTIFACT
pip install agent-trace-card
Portable cards for one run.
Portable Markdown and JSON cards for one agent run: tools, retries, data touched, outcome, failure mode.
LICENSE
MIT · public
OTEL-EVAL-BRIDGE
№ 05
ARTIFACT
pip install otel-eval-bridge
OTEL spans into eval cases.
Bridge OpenTelemetry and Phoenix GenAI spans into redacted eval regression cases.
LICENSE
MIT · public
PROMPT-RUBRIC-DRIFT
№ 06
ARTIFACT
pip install prompt-rubric-drift
Drift notes for PR review.
No-model PR review notes for prompt and rubric changes: weights, criteria, boundaries, injection exposure.
LICENSE
MIT · public
ROBOTICS DATA · 4 packages
LeRobot quality gates, recovery and intervention metrics, VLA robustness probes, and embodiment release cards.
LEROBOT-QUALITY-GATES
№ 07
ARTIFACT
pip install lerobot-quality-gates
LeRobot dataset gates.
Local quality gates for LeRobot-style datasets: metadata, episodes, sensors, action and state fields.
LICENSE
MIT · public
ROBOT-RECOVERY-BENCH
№ 08
ARTIFACT
pip install robot-recovery-bench
Recovery and intervention metrics.
Schema and metrics for human intervention and recovery segments, including repeated-failure clusters.
LICENSE
MIT · public
VLA-ROBUSTNESS-KIT
№ 09
ARTIFACT
pip install vla-robustness-kit
VLA perturbation probes.
Simulator-light VLA diagnostics for language, vision, metadata, task-phase, and embodiment perturbations.
LICENSE
MIT · public
EMBODIMENT-CARD
№ 10
ARTIFACT
pip install embodiment-card
Release cards for robot data.
Structured robot dataset and VLA release cards for sensors, action spaces, frames, control rate, limits.
LICENSE
MIT · public
REVIEW SURFACE · 2 packages
Synthetic failure cases with reproducible commands. Outreach packets that carry no endorsement claim.
FAILURE-GALLERY
№ 11
ARTIFACT
pip install failure-gallery
Reproducible failure cases.
Synthetic agent and robotics failure cases with reproducible commands and expected review labels.
LICENSE
MIT · public
LAB-OUTREACH-KIT
№ 12
ARTIFACT
Repo only · no PyPI rollout yet
Packets for lab review.
Technical review packets, no-endorsement language, and a feedback-log schema for lab review asks.
LICENSE
MIT · public
OPEN V2
Useful. Public. Careful about claims.
Practical trust infrastructure. Not certification. Not endorsement. Read the source. Run it locally.