Undo for AI agents: a draft spec for taking back what an agent already did

A developer has published a draft specification, HATP v0.3, proposing a "Journal" mechanism that gives AI agents an undo capability for actions already taken. A Recorder sitting between the agent and its MCP tool servers captures each tool result with a change summary, a compensation token and an expiry window, chaining entries into a signed, hash-chained Run Record; tool servers without HATP support are logged as irreversible. Each operation declares whether it can be undone and at what cost or side effect, letting organizations allow reversible actions freely while routing irreversible ones through human approval.

A sales agent is tidying up the Q4 pipeline. It marks deal D-881 as Won in the CRM, books a kickoff meeting with the customer, and emails their buyer a welcome note. An hour later you notice it picked the wrong deal. The customer hasn't signed. To clean up, you open three systems, work out what changed, put back the deal stage unless a colleague has edited it since , cancel the meeting which tells the attendees , and accept that the email has gone. In my last post https://dev.to/murthyg/human-approval-as-a-2pc-vote-a-draft-protocol-for-safe-multi-tool-agent-actions-1j05 I described HATP , a draft protocol that makes multi-step agent actions all-or-nothing before they happen: plan, prepare, get a human's signed approval, then commit. Your comments on that post shaped most of v0.2, so thank you. But most agents don't plan. They call one tool at a time, and you find out afterwards. So for v0.3 I've drafted the other half: the Journal , undo for agents. Nothing is built yet. I'd like your opinion before we start. | What exists | What it does | What's missing | |---|---|---| | MCP tool annotations destructiveHint , idempotentHint | Tell the client a tool might be dangerous | Hints only: no way to reverse anything | | Sagas and compensating transactions | Undo steps inside one application | Nothing that works across independent tool servers an agent calls | | HATP v0.1 compensate tier | Undoes steps when a transaction fails partway | Only inside a planned transaction, and only before it succeeds | | Agent traces and logs | Show what happened | Can't take anything back | A Recorder sits between the agent and its tool servers, usually an MCP gateway. The agent needs no changes. When a tool server supports the journal, its normal tool result carries three extra things: what changed, a token to reverse it, and how long reversal stays possible. " meta": { "hatp/journal": { "undo": { "kind": "compensate", "undoBy": "2026-11-07T14:02:10Z" }, "compensationToken": "ctk 9QwX…", "effect": { "summary": "Deal D-881 stage: Proposal → Won", "resource": "crm://acme/deal/D-881", "version": "v13" }, "receipt": { "…": "signed by the CRM" } } } The tool server writes the summary from the actual change, not from the agent's text, because a human reads it later when deciding what to undo. The Recorder chains every entry into a signed, hash-chained Run Record . Tool servers that know nothing about HATP still get journaled. Their calls are recorded as unknown and treated as irreversible. So it works with any MCP server on day one, and gets better as servers add undo. Each operation declares up front whether it can be undone: | Operation | Declares | Why | |---|---|---| | update deal | compensate , 7 days | Old stage can be restored | | create event | compensate , 30 days, side effect: "attendees get a cancellation notice" | Undo is possible but visible | | refund order | compensate , 1 day, cost $2.50 | Undo has a fee | | send email | none : "delivered to external recipients" | Can't unsend | | export report | none : "discloses data outside the organisation" | A read you can't take back | Once every operation says whether it can be undone, you can stop asking "should this agent be allowed to act?" and ask "can we take it back if it's wrong?" { "autonomy": { "compensate": "allow", "unknown": "ask", "none": "ask", "approver": "human://acme.com/role/team-lead" } } The agent acts freely on anything reversible. Before anything irreversible, the Recorder doesn't forward the call: it runs it as a normal HATP transaction with a human approval, so the email goes out only after someone signs off. Your organisation sets this policy, not the agent, and the agent can't change it. Reversing work can destroy someone's newer edits or notify people outside the company, so a reversal goes through approval like any other risky action. You ask to undo the run "from step 2 onwards". The Recorder then: Undo 2 actions by sales-bot for Dana: 1. Cancel "Kickoff with Globex" on Nov 4 attendees get a cancellation notice 2. Deal D-881 stage: Won → Proposal Cannot be undone: - Email to buyer@globex.com delivered to external recipients For what can't be undone, it suggests a remedy, such as "send a correction email", but never performs it. A corrective action is a new action and needs fresh consent. Between the agent's change and your undo, Eve may have moved the deal to "Negotiation". Blindly restoring "Proposal" would wipe out her change. So every compensation carries ifVersion : the version the agent left behind v13 . If the record has changed since, the tool server refuses with UNDO CONFLICT and says what changed. The approval screen then shows the conflict, and only a human can choose to overwrite it. HATP v0.1 compensates only when a transaction fails partway. With the Journal, a coordinator records a committed transaction as a run, so the same tokens can reverse it later: "that trip committed fine, but it was the wrong week." compensationToken and undoBy from your MCP server's tools? prepare . check instead of the full two-phase flow; only the human votes. Is that too weak? Spec and SDK: https://github.com/probita-ai/hatp https://github.com/probita-ai/hatp . Comments here or issues on GitHub are both welcome.