Harness Hub
All open source

Sponsio

Deterministic safety solutions for probabilistic AI agents.

SponsioLabs/Sponsio

Maintained by SponsioLabs

Open-source project described as deterministic safety solutions for probabilistic AI agents. Topics include guardrails, a policy engine, intent verification, runtime safety, and prompt injection. Packaged as a skill for OpenClaw.

Activity

Last push
· yesterday
Latest release
v0.2.0a9
· 2mo ago
Last 90 days
14 releases
  1. v0.2.0a21Pre-release

    · 3w ago

    Upgrade with pip install --pre -U "sponsio==0.2.0a21", then re-run sponsio host install and sponsio doctor --tier. This release lets Sponsio run out of the agent's reach. Tier 2 (sudo sponsio install --system) runs the daemon as its own user behind root-owned hooks. Tier 3…

  2. v0.2.0a20Pre-release

    · 3w ago

    Security: this release fixes ways an agent guarded through the host hooks could disable or weaken enforcement in 0.2.0a19 and earlier. See GHSA-xh83-347v-qrfw. Upgrade with pip install --pre -U "sponsio==0.2.0a20", then re-run sponsio host install and sponsio doctor. Added…

  3. v0.2.0a19Pre-release

    · 5w ago

    Follow-ups to the kta1kri rounds (#180, #188, #196), found while re-verifying 0.2.0a18 through the real install path. TypeScript @sponsio/sdk@0.2.0-alpha.9 carries the same fixes. A rule written in YAML/LTL binds every spelling of its tool a17 canonicalised the tool argument…

  4. v0.2.0a18Pre-release

    · 5w ago

    A second security release from the kta1kri report (thanks again). The reporter retested 0.2.0a17 through the real sponsio host guard claude-code --stdin path and found three findings not fully closed. If you run the Claude Code hook, upgrade. TypeScript…

  5. v0.2.0a17Pre-release

    · 6w ago

    A security release. It closes an external report (thanks to kta1kri) and the follow-ups that finishing it turned up. If you enforce contracts in production, upgrade. TypeScript @sponsio/sdk@0.2.0-alpha.7 carries the same fixes. A rule that cannot be evaluated no longer counts…

  6. v0.2.0a9v0.2.0a9 — a project's second agent works

    · 2mo ago

    Follow-up to 0.2.0a8. All of it was found by running a real multi-agent app against a console that already held other agents' books — the configuration nobody tests on a fresh install. Added An agent's first run puts its book in the cloud. Nothing pushed before: the cloud…

Checked against GitHub on Nov 21, 2026. Release notes are shortened to plain text.

ECC

affaan-m/ECC

Agent harness performance optimization system: skills, instincts, memory and security for coding agents.

Maintained by Affaan Mustafa

4 pros · 4 cons
Featured

NousResearch/hermes-agent

The agent that grows with you.

Maintained by Nous Research

v0.23.3released
4 pros · 4 cons

JuliusBrussee/caveman

A token-saving skill plus proxy that makes coding agents talk tersely.

Maintained by JuliusBrussee

v4.1.0released
4 pros · 4 cons

Connections

Relations from the Harness Hub knowledge graph, as charted in the State of Harness reports.

Maintained by · 1

  • SponsioLabsOrganisation

Built on · 1

Sources

DisclosureAdded by Harness Hub's automated review of public metadata; not independently tested.

  1. GitHub repo

The Harness Hub newsletter

New harness companies, projects, releases and the monthly State of Harness report, in your inbox.