Launching today
Slipmark
Make AI agents prove what they checked before anything ships
1 follower
Make AI agents prove what they checked before anything ships
1 follower
Make AI agents prove what they checked. Bind every agent verdict to a SHA-256 of the files it read, and refuse to ship on mismatch.
Free Options
Launch Team
Render Sandboxes: a computer for every agentRun untrusted code safely. Idle sandboxes cost nothing
Promoted
I built Slipmark because my own nightly agent pipeline lied to me.
A security audit gate reported green every night for months. It was inspecting the wrong branch. Nothing checked that the verdict matched the code about to ship.
Slipmark fixes that. Every agent verdict is bound to a SHA-256 of the exact files the agent read. Before anything ships, the gate re-hashes and refuses on mismatch. When I pointed it at 23 open agent PRs, all 23 failed.
Open source, Apache-2.0, runs as a GitHub Action. Also listed in the MCP Registry. Slipmark Cloud adds a public ledger and audit evidence packs if you need them.
Feedback I'd most value: how you run agents in CI today, and what evidence you'd need before trusting an agent's approval.