I tried to forge my own AI agent's audit log

A developer built an open-source audit and governance layer for AI agents, comprising a versioned "passport" identity card, a scoped machine-readable mandate, a pre-model-call permission gate, and an append-only hash-chained JSONL journal. Testing the journal by tampering with a record — flipping a "deny" decision to "allow" — the verification tool detected the edit and named the specific record, demonstrating tamper-evident logging. The work responds to Cloud Security Alliance and 2026 State of AI Agent Security findings that nearly half of production agents lack monitoring and roughly a fifth have their own identity.

Half the organisations running AI agents have already had one step outside its permissions. Nearly half of the agents in production have no monitoring at all. Only about a fifth treat an agent as something with an identity of its own. Those are numbers from the Cloud Security Alliance and the 2026 State of AI Agent Security report, and they describe the same gap from three directions: agents act, and nobody can say afterwards what exactly they did. I spent a few weeks building the missing piece, and then spent an afternoon attacking it. This is what survived. A passport. Not a config file — a card for one version of an agent that says who built it, who runs it, what it does, and what it will never do: { "agent id": "kepil.leads.v1", "purpose": "Handle inbound requests and hand qualified ones to a person", "does not": "never promises prices or deadlines on the company's behalf", "never sends invoices or contracts", "never exports the customer base outside the perimeter" , "risk class": "medium", "autonomy class": "medium", "risk review": { "last": "2026-09-14", "next due": "2027-09-14" } } Trimmed for readability — a real card also carries the operator, the model list, the rationale for each class, an incident log and the passport's own hash. One caveat before you install: the runtime, the panel and these field values are currently in Russian. The format and the code are not language-specific, but you will be reading «средний» where this article says "medium". A new version is a new card. The old one is kept forever, because the question "what was this agent allowed to do in March" has to have an answer. A mandate. A machine-readable power of attorney for one job — not a permanent grant: { "mandate id": "mnd-0001", "allowed actions": "read:inbox", "read:crm", "generate:reply", "send:message", "write:crm", "generate:summary", "send:handoff" , "allowed systems": "crm.local", "whatsapp.local" , "forbidden actions": "sign: ", "pay: ", "export:database", "send:bulk" , "human confirmation required": "send: ", "write:crm", "publish: " , "limits": { "messages": 300, "llm cost kzt": 4000 }, "valid until": "2026-09-21T10:09:45" } Anything not explicitly allowed is refused. That sounds obvious until you look at how agents are actually wired: one shared API key, full network access, and a prompt asking nicely. A gate. One place an agent touches the outside world, and every action is checked before a model is even called. The checks run in a fixed order: passport active, mandate valid, action allowed, system allowed, limits not spent, irreversible or not. Any error inside the check means refusal — never a pass. A false refusal costs a retry; a false pass costs a breach notification. A journal. Append-only JSONL where each record carries the hash of the one before it: {"seq":10241,"ts":"2026-09-12T11:42:07+05:00","agent id":"kepil.leads.v1", "action":{"type":"send:message","target":"whatsapp.local"}, "decision":"await human","cost kzt":0, "prev hash":"sha256:c61d8b…","hash":"sha256:d5ade5…"} Values of personal data never enter it. Types, counts and hashes do. A log you cannot verify is a diary. So I wrote a record, then went back and edited it the way someone covering their tracks would — flipping a refusal into an approval: bash the agent tried to sign a document; the gate refused $ npx proofbyte-agent-trace verify journal.jsonl Integrity confirmed. Records: 4 Chain head: sha256:d58c204ff2a37b093307b95b3d9fcc4edc510cd6062622db10b3461a4341dd3f $ sed -i 's/"decision": "deny"/"decision": "allow"/' journal.jsonl $ npx proofbyte-agent-trace verify journal.jsonl Integrity BROKEN. Record 3: record content changed after it was written Two things matter here. The first is that it names the record, not just the file. Hash chaining gives you that for free, and it is the difference between "something is wrong" and "here is what was changed". The second is subtler. The verifier is a different program in a different language, written by nobody in particular. The writer is Python; the verifier 0.0 as "0.0" and JavaScript as "0" , so the same record hashed Kepil ships an MCP server, a JSON API and an n8n node. All three can create orders, run steps, ask whether an action is permitted, and read the journal. None of them can confirm. If a model could call confirm , the human would drop out of the chain: the agent would be approving its own irreversible actions, and every guarantee above would become decoration. So the confirmation card goes to a person — the panel, or Telegram with two buttons — and there is a test whose only job is to fail if confirm ever appears in the tool list. I find this the most useful thing to say about the whole project. Everyone advertises what their agent framework can do. The interesting part is what it refuses to do even when you ask it nicely. Recording is table stakes. Agent-governance products — Workday's Agent Passport, Okta, Credo AI — all write logs. None of them put anything back. Because the journal is a graph of actions and every profession declares its compensating action, you can walk it backwards: "rollback": { "send:message": "send a correcting message", "write:crm": "restore the previous state of the deal from the journal", "publish:complaint": "" } The empty string is the important entry. It means this cannot be undone . Pick a window in the panel — the last hour, say — and the pass runs from the most recent action backwards, applying each compensating action in turn, and stops at the first step whose compensation is that empty string. It then reports both halves: what it undid, and what it could not. The panel shows the same thing before you press the button, naming the step where the pass will stop: Will undo 2 actions. The pass will stop at "Search for tailored requirements" — that action is irreversible, and nothing before it will be undone. An undo promise that quietly fails is worse than no undo at all, so there is a test whose only job is to fail if the preview and the engine ever disagree. One more thing follows from this. A rollback is written to the journal as an operator's decision, so neither the MCP server nor the JSON API can perform one — same reasoning as confirm . An agent undoing its own actions would be signing in somebody else's name. The core has zero dependencies. Not "few" — zero: it is the Python 3.11+ standard library, and CI fails the build if a third-party import appears. Two reasons. It installs inside an air-gapped perimeter where pip install from the internet is not an option. And a tool that sees every action an agent takes should not drag a hundred transitive packages behind it. State is JSON files. No database. You can open them, read them, and attach them to a dispute — which is the whole point of the exercise. Alpha, 115 tests, AGPL-3.0. It does not run agents for you: it is the layer that says what they may do and records what they did. pip install kepil python -m kepil.admin If you are running agents in production and can answer "what did agent X do on 9 September, and who allowed it" — I would genuinely like to know how. If you cannot, that is the gap this was built for.