Afterstate Action integrity

Research

Findings

Notes from the proving ground — integrity failures that transcript-only evals miss, and what changes when agents reconcile against the system of record.

  1. 01

    Claim

    Before the write runs, lock a business-operation identity — not a random retry id.

  2. 02

    Execute once

    The tool runs. If the response is lost, the outcome is marked unknown — never assumed.

  3. 03

    Observe

    Before any retry, check the system of record. Evidence, not a guess.

  4. 04

    Receipt

    applied / not_applied / ambiguous, plus the next step. Every study below grades against this.

Integrity study

False failures on τ³-bench: when the cancel worked and the agent said it didn’t

Fault-injected grading on Sierra τ³-bench write tools. Scale study: 2,016 traces — afterstate 100% safe and complete; timeout slice still shows naive at 12% verified completion.

Read the study →
Adversarial study

Where Afterstate breaks

780-cell adversarial matrix. Transport faults: 100% safe. Identity collision (deliberate misconfig): 0% safe. Observer lag: safe but incomplete. Blended afterstate safety: 92% — and we published why.

Read the boundary →
Integration demo

QuickServe × Afterstate — integrate in four steps

Simulated design partner. Before: 10 duplicate writes, $85.60 double-processed. After: 0 duplicates. Real init --apply, verify scorecard, and receipt log.

See the dashboard →
Research note

Transport retries are solved. Agent retries are not.

Primary-source evidence that agent-level re-invocation sits above transport idempotency — and why observe-before-retry belongs in the write path.

Read the note →