Store, measured at rest
Atelya Attest
keyed hash chain, verify_chain audit
Atelya Attest was seeded through its own API and edited behind its back nine ways. 2 edits served as genuine, 7 reported on audit, 0 rejected on read.
- Measured on
- atelya-attest 0.1.1, Darwin arm64, Python 3.12
- Date
- Source
- github.com/RonaldSit/atelya
- Attack versions
- tamper@v1, truncate@v1, delete_middle@v1, reorder@v1, forge@v1, cross_replay@v1, rollback_replay@v1, metadata_tamper@v1, snapshot_rollback@v1
What this means
The agent acts on an edited memory first; the operator finds out only if the audit is run. 7 of the nine edits are reported by that audit and 2 are served with no record anywhere. Run the audit on a schedule, or move the check onto the read path.
The 9 verdicts
Detection point: the audit, a separate call the operator has to make; the read path serves the edit first.
Level L0, Measured. The store has a published row. Any verdicts. What L0 means.
| Edit | Verdict | What the tool said |
|---|---|---|
| T1 Content tamper | reported | detected on reload (verify_chain: seq=2: curr_hash mismatch (payload modified after attestation)) |
| T2 Tail truncation | accepted | accepted silently |
| T3 Middle deletion | reported | detected on reload (verify_chain: seq=2: sequence gap/reorder: entry has seq=3, expected 2) |
| T4 Reordering | reported | detected on reload (verify_chain: seq=1: sequence gap/reorder: entry has seq=2, expected 1) |
| T5 Forged insertion | reported | detected on reload (verify_chain: seq=5: prev_hash does not match previous curr_hash (deletion/reorder)) |
| T6 Cross-context replay | reported | detected on reload (verify_chain: seq=4: prev_hash does not match previous curr_hash (deletion/reorder)) |
| T7 Rollback replay | reported | detected on reload (verify_chain: seq=4: sequence gap/reorder: entry has seq=0, expected 4) |
| T8 Metadata tamper | reported | detected on reload (verify_chain: seq=2: curr_hash mismatch (payload modified after attestation)) |
| T9 Snapshot rollback | accepted | the older copy opened as current; the newest genuine record is gone without an error |
A verdict is what the tool did, not an opinion. "Accepted" means it loaded the altered store, raised nothing, and the agent carried on from the altered memory as if it were true. Every cell has a control that proves the edit landed before the verdict counts.
Where the attacker stands
The attacker holds the store (a file, a table, a bucket, or the data-plane role of a managed service) and edits it outside the tool's API, then the tool is reopened the way its users would reopen it.
Reproduce this row
Everything runs offline unless the store is a managed cloud service, in which case the row needs a project of your own. The run seeds a fresh store, applies each edit, confirms it landed, reopens the store and records what came back.
pip install agent-memory-integrity
python agmi/full_runner.py --json results/scorecard.json # every row, this one includedBadge
Maintainers can link their row from their README. The badge points here and changes nothing on your side:
[](https://agentmemoryintegrity.org/stores/atelya-attest-chain.html)