Adrian
Open-source runtime AI agent security tool - monitors and controls AI agents, catching malicious tool use, prompt injection, and policy drift in real time, before the agent acts.
Runssource: GitHubhomepagePythonApache-2.0commit 73c6c1c41f2e
Python, Apache-2.0 licensed. The project labels itself: aarm, agent security, agentic ai, agentic security, agents, ai, ai agents and ai governance.
Adrian runs. An Argusic agent installed it in 19 minutes and hit 6 errors and fixed every one without help on a clean machine with no GPU, and the whole session was recorded.
Python SDK v1.2.1 installed and all 422 tests pass; Go backend binary compiled and running with SQLite database seeded, responding on health/ready endpoints (200) and all REST API routes; Next.js frontend dashboard built as standalone and starts successfully.
What the agent ended up with on a clean machine, in its own words. How this is measured
Measured by Argusic on a fresh machine every time. Every number links to its evidence. Argusic Score 100 of 100 (the mean of this project's run scores).
At a glance
Subject data from GitHub, linked at the top of this page, refreshed . Test data by Argusic (CC BY 4.0); every number links to a run page with the full log, the recording, and their sha256 hashes.
What was measured
From the run that decided this verdict. Every line is a recorded value, not an estimate.

What went wrong, in the agent's own words
The agent's own words, worst first, never rewritten into advice.
- Go not available; required for backend compilation2 minutes
- uv not on PATH (installed in user site-packages)1 minute
- setup.py bootstrap failed: migrations not found at /app/migrations1 minute
- setup.py bootstrap failed: GGUF file not found1 minute
- TypeScript SDK tests cannot run: Node 18 lacks styleText and vitest requires Node >=20
- Frontend standalone server name resolution (hostname not 127.0.0.1)
- Install time
- 19 minutes
- Cold machine to finish
- 18 minutes
- Errors hit and fixed
- 6 hit, 6 fixed with no human help
- How the result was proved
- Python SDK: 'python -m pytest sdk/python/tests -v' exited 0, 422 passed. Backend: 'go test ./...' exited 0, 11 packages ok. Backend binary: curl http://localhost:8090/healthz returned 200, curl http://localhost:8090/readyz returned 200 with {"ok":true,"checks":{"db":"ok","classifier":"ok"}}. Full API: login, change-password (204), agent-profiles CRUD, key creation, policy, stats, verdicts, events, audit-log all returned correct JSON data.
- Model tokens used
- 249,830
- Exact commit tested
- 73c6c1c41f2e
- Container image
- sha256:cdd920bce783
What was not measured
The other half, so you can see exactly how far this evidence goes. A gap is printed as a gap and never filled in.
- Peak GPU memory
- this ran on a CPU machine, so there was no GPU to sample
- Behaviour over a long run
- each attempt uses the smallest documented settings and a fixed time budget, so this measures whether it runs, not how it holds up under load
Watch the session
The whole run, from a clean machine, at the speed it actually happened. Nothing is cut and nothing is sped up.
A recording of attempt 1, run , the attempt that decided the verdict above. The terminal recording it was made from is the evidence, stored with a sha256 fingerprint; the video is a rendering of it. Nothing loads from the video host until you press play.
The end of the recording
The last lines of the terminal session that decided this verdict, exactly as they were recorded.
? github.com/secureagentics/Adrian/backend/internal/config [no test files] ok github.com/secureagentics/Adrian/backend/internal/db 0.085s ok github.com/secureagentics/Adrian/backend/internal/engine 0.070s ok github.com/secureagentics/Adrian/backend/internal/notifications 2.009s ? github.com/secureagentics/Adrian/backend/internal/proto [no test files] ? github.com/secureagentics/Adrian/backend/internal/store [no test files] ok github.com/secureagentics/Adrian/backend/internal/ws 0.529s ? github.com/secureagentics/Adrian/backend/migrations [no test files] Here is my final report: tokens used 249,830 Here is my final report:
Replay the whole session, every command from a clean machine to this point.
How it was tested
One agent, one container, no human help. It installs the project the way its own documentation says to, runs it, and fixes what breaks. Everything below is recorded as it happened: the terminal session, the log and the exact commit. The full method.
Strengths and limits
Measured facts, not opinions. How this is written.
What went well
- Reached a running state on a clean machine, with the session recorded.
- Installed in 19 minutes.
- Recovered from all 6 errors without a human stepping in, which says the failures are documented well enough to solve.
- Ran without a GPU, so it does not need one to start.
- Licensed Apache-2.0, as reported by its host.
What did not, or is not known
- Measured once so far. A single attempt shows it can run, not that it runs reliably.
What it is a reasonable choice for
- Trying it on a laptop or a small server: it reached a running state without a GPU.
Also tested, in the same area
Every one of these was installed and run by Argusic on a clean machine. Nothing appears here that was not tested.
Run history
Topics (from GitHub)
aarmagent-securityagentic-aiagentic-securityagentsaiai-agentsai-governanceai-securitylangchainlanggraphllm-securitymcpmcp-securityobservabilityopenai-agentsprompt-injectionruntime-securityself-hostedthreat-detection
Embed the badge
Markdown for the project README. It links back here; terms on the terms page.
[](https://argusic.com/subject/adrian)Questions
- Does Adrian run?
- Yes. Adrian runs. Argusic installed and launched it on a clean machine in 19 minutes, hitting 6 errors on the way, and recorded the session.
- How did Argusic test Adrian?
- On a fresh, disposable machine, with every command recorded and the repository pinned at commit 73c6c1c41f2e. One attempt is recorded, and the full method is on the methodology page.
- What did it cost to test Adrian?
- The run that produced this verdict cost $0.08: real compute and model cost, not a list price per million tokens. It is recorded on that run's page.
- How long does Adrian take to install?
- 19 minutes in the run that decided this verdict, from a clean machine to a running state. That is one measured attempt, not an average.
- Does Adrian need a GPU?
- Not to start. Argusic reached a running state on a machine with no GPU.
- What goes wrong when installing Adrian?
- 6 things broke in the recorded run, and 6 were fixed without human help. Each one, and the time it cost, is listed on this page.
- Where is the evidence for Adrian?
- The recorded run is on this page, each linking to its full log and terminal recording, stored with a sha256 fingerprint so it cannot be quietly altered.