Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
29 changes: 29 additions & 0 deletions docs/audit-evidence.md
Original file line number Diff line number Diff line change
Expand Up @@ -636,3 +636,32 @@ the previous `evidence_hash` stays readable in the history of each `submit_verdi
the report that hashes to it is the `reports/` file at the commit before this run. Only
`price-checker` and `cctp` change verdict. The rewritten `reports/` files are committed with the run log, and the API
redeployed so `/reports` serves the bytes the chain now anchors.

### Executed, 17 September 2026

Run from `main` at `5a938f5` (PR #43 merged), against Registry v2 and Tokens v2, with the command
above. Record: [`evidence/ste-37-chain-correction-2026-09-17.json`](evidence/ste-37-chain-correction-2026-09-17.json)
(journal alongside it). **26 on chain, 0 held back, 0 skipped, 0 failed.**

| skill_id | version | before (15 Sep) | after | `submit_verdict` | `mint_verified` |
|---|---|---|---|---|---|
| `com.fixtures.safe.price-checker` | 0.9.0 | DANGEROUS 10 ([`75055115…`](https://stellar.expert/explorer/testnet/tx/75055115da6c6e4ea4aba67a51589ff6dc741ed62c84ebefcf854f214ad4d8b1)) | **SAFE 90** | [`3bd64c68d2231c8f…`](https://stellar.expert/explorer/testnet/tx/3bd64c68d2231c8fb887bc3fc4b0196b49c680688e03dc71a86c4e4f170ab3a3) | [`279a038847189dd2…`](https://stellar.expert/explorer/testnet/tx/279a038847189dd2c5b3635724393a877c801f3dbf06a57a812729e19fe48faf) |
| `org.stellar.skills.cross-chain.cctp` | 2026.8.31 | DANGEROUS 10 ([`542468dc…`](https://stellar.expert/explorer/testnet/tx/542468dcf909f46111f0a41b207819ef01f1abad6501b270f9879a9ba96368db)) | **SAFE 100** | [`3490559a37fa8c2b…`](https://stellar.expert/explorer/testnet/tx/3490559a37fa8c2b54e5acbb48064a5172fce4a1e315f924fae0c1abf767df22) | [`9ecc990a69df512c…`](https://stellar.expert/explorer/testnet/tx/9ecc990a69df512c272d0cd896952be2f2269416dd3dced2783666f019caec61) |

- **23 hash-only re-anchors**, as the dry run said: same verdict, same score, new `evidence_hash`
(the STE-39 byte shift described above). Each one's `submit_verdict` hash is in the record.
- **5 rows DANGEROUS on purpose**, each marked `dangerous_intended` in the run log: the four
`com.fixtures.poisoned.*` fixtures and `com.fixtures.demo.release-notes` 2.0.0 (the rug pull).
All are ours and all carry `expected_verdict: DANGEROUS` in the corpus.
- `com.fixtures.demo.changelog-writer` 2.0.0 stays register-only and `UNAUDITED`; nothing was sent.

**Checked afterwards by reading the chain, not the run log:** all 25 audited versions read back
from Registry v2 with the verdict and score in the record and an `evidence_hash` equal to
`sha256` of the committed `reports/` file; both corrected versions answer `is_verified: true`.
Every anchored hash is also **identical to the dry run's**, byte for byte.

One thing that differed from the dry run and is worth stating: the environment of the real run
had a model key set, so stage 3 called the model for each entry. Since STE-39 the model is
advisory and its trail is excluded from the hashed report, and the identical hashes are the
evidence that this now holds in practice, not only in the unit tests.

Loading