The pre-deploy release gate for AI agents. Paste any repo — release-gate finds the agent-layer risks SAST, guardrails, and evaluators structurally miss: model output reaching eval (the CVE-2025-51472 RCE class), user input in a system prompt, LLM loops with no cost ceiling. Every finding cites its evidence. Takes 10 seconds.
A SAST tool sees eval(x) but can't tell that x came from the model. That blind spot is the whole agent layer — and it's the gate release-gate stands on.
🔒 Scanning a private repo? Install the release-gate-ai App to grant read access — public repos work without it.
Running audit…
Sample output
One of two axes. Every safeguard maps to a real production failure — no checkbox theater. The other axis, Agent Code Safety, scans the code itself for the agent-layer risks SAST misses (below).
Governance lives in heads, not docs. Knowledge walks out the door.
No tests = no signal. You're guessing in prod.
Silent tool misuse. No audit trail when something goes wrong.
One prompt injection → $10k overnight. No ceiling = no floor.
Agent misbehaves at 2am. No kill switch = 45-min deploy to fix it.
Incident at 3am. Who do you page? "The team" is not an answer.
Unprotected endpoint = denial-of-wallet attack in minutes.
Agent Safety Scan
Point it at any GitHub repo running an LLM agent. It finds the failure modes that only exist once a model is in the loop — prompt-injection surfaces, exec sinks fed by model output, LLM calls with no token ceiling, and hardcoded keys. SonarQube won’t flag these; this is the layer on top.
github.com/your-org/your-agent. Sign up free to see every finding with file:line.
Who it's for
Whether you're a solo engineer or a platform team governing dozens of agents — release-gate fits into how you already work.
One pip install. Works in your terminal, your CI, or via API. No cloud containers, no agents to host, no SaaS dependency for the core tool.
Your machine. Your terminal. No account needed.
5 lines in your workflow. Score appears in PR summary.
Use from any language. Full history + dashboard for teams.
Every agent loop needs a checker that isn’t the same model that wrote the output. Release Gate owns the Verify step — returning CONTINUE, SHIP, or ROLLBACK after each iteration.
Works in any shell loop. No account needed.
Call from any language. Each iteration is persisted.
Declare the policy once. Release Gate enforces it every iteration.
/api/verify endpoint — requires a free account and API token.
Without auth the result is a local simulation.
Three lenses on the same agent. All run server-side and return one decision.
The CLI is always 100% free and unlimited — no scan caps, no account required, works in any CI pipeline. The platform adds what CLI can’t: persistent history, team visibility, and compliance audit trail.
Scan limits below apply only to hosted web scans. Local CLI and CI usage is always unlimited.
Why pay if CLI is free? History & full results in the browser.
Why pay if CLI is free? Score trends, team API tokens, email alerts when score drops.
Why pay if CLI is free? Org-wide governance policy, compliance audit trail, SSO.
Fits your stack
SonarQube, Snyk and your SAST suite keep your code healthy. release-gate keeps your agent safe to ship. They’re complementary — release-gate doesn’t replace anything you already trust, it covers the layer those tools were never built to see.
eval/exec/shell sinksA clean SonarQube report means your code won’t leak a buffer or run an injected query. It says nothing about whether your agent has a budget ceiling, a kill switch, or a prompt-injection surface in its system message. release-gate closes that gap — so “the code passed” and “the agent is safe to deploy” finally mean the same thing.
We audit your repo — we don’t store your code. Here’s exactly how access works.
FAQ
What release-gate is, what it tests, and where it fits next to the tools you already use.
py:), a shell command (cmd:), or any HTTP endpoint. For HTTP you just tell it where the input and output live in the JSON — so LangServe, a plain FastAPI route, or an OpenAI-compatible API all work with no wrapper: agent-score "http://host/run#in=prompt&out=reply".exec/shell sinks fed by model output, LLM calls with no token ceiling, hardcoded keys. It moves per repo and doesn't depend on adopting anything from us. Governance is maturity: have you declared the enforceable safeguards (budget ceiling, kill switch, owner, evals, trace policy)? A low Governance score means undeclared, not unsafe. Keeping them separate is why a clean codebase with no config still scores well on safety — no circular "you didn't adopt our file" penalty. Each score also shows the findings driving it, so it's never a black box.exec/a shell, an uncapped loop that can burn $10k overnight, a missing kill switch. SonarQube has no concept of any of those. Keep your SAST suite — this is the agent layer on top, not a replacement.max_tokens or budget. The pattern behind the $4k weekend bills and $6.5k runaway-agent stories.while True: with no iteration cap and a stop condition that may never trigger. The AutoGPT-style runaway that turns a $5 task into $400.eval(), exec(), shell=True, new Function() fed by model or user input — a remote-code-execution path.client.chat.completions.create(...) without a ceiling but not a harmless wizard.run().agent-score or the on-site live scan), it runs four weighted dimensions. Safety plants a canary secret in the agent's context and runs a tiered prompt-injection / exfiltration battery (L1 direct → L4 multi-turn); a confirmed leak is a hard BLOCK. Correctness runs your domain evals. Loop behavior checks convergence and catches runaways. Cost & latency characterizes per-call spend. A high total can't buy back a weak dimension — promote floors keep a broken-but-safe agent out of PROMOTE.audit <repo>) is free and makes no model calls.NOT_ASSESSED where it doesn't test something, because a fake green fails an audit harder than an admitted gap.Sign up (free) to send us your question — that way we have an email to reply to.
No sign-up required for the first scan.
All repos, live risks, and historical trend in one view.
Loading…
Loading report…