10 projects
stataudit
Check AI-written statistics with evidence: re-run the analysis, audit each test's assumptions on the real data, recompute reported p-values, lint analysis code, and verify proposed fixes.
proofcheck-ai
Evidence, not authorship: check whether a pull request actually does what it claims.
numguard-ci
Numerical edge-case testing for array code: catch the NaNs, overflows and precision bugs that happy-path tests miss, in pytest and CI, with an optional agent whose fixes are verified by execution.
pytest-numguard
pytest plugin for numerical edge-case testing: NaN, inf, overflow, cancellation and dtype bugs in array code (installs numguard-ci).
provio-sdk
Policy-gated tool calls for Python agents: every tool call is checked against provio.yaml and recorded to provio's hash-chained ledger.
provio
A safety floor your AI agents can't get under: one policy for Claude Code, Codex, Gemini CLI, Cursor, Windsurf and MCP, judged by what each call will run, with a tamper-evident, signed ledger.
writ-sdk
Renamed to provio-sdk: pip install provio-sdk. This package installs provio-sdk and keeps `import writ_sdk` working.
writ-cli
Renamed to provio: pip install provio. This package installs provio and keeps a `writ` command that runs it.
rollback-mcp
A reversibility layer for agentic tool use — a transparent MCP proxy that classifies which tool calls can be taken back, synthesises inverse operations, keeps a durable cross-server undo log, and interrupts you only for genuinely irreversible actions.
verifydoc
A trust layer for document -> structured-JSON extraction: calibrated per-field confidence, source grounding, and abstention on top of any extractor.
