Verify a decision
Every moderation decision on AVOID.NET is anchored to the Solana blockchain. You don't have to trust us — you can verify cryptographically that we committed to a verdict at a specific moment and have not rewritten it.
How verification works
- We commit. When a moderator accepts/rejects a submission, we serialize the decision into deterministic UTF-8 bytes (
payload_canonical_string), hash it with SHA-256, encode the digest as base58, and write it to Solana inside an SPL Memo v2 transaction. - We store the bytes. The exact bytes we hashed are stored alongside the decision in our database. Anyone can read them and recompute the hash in any language.
- You compare three values. Database hash, your independently-recomputed hash, and the hash inside the on-chain memo. If all three match, the decision is authentic and timestamped.
The on-chain memo format is
AVOID.NET|v1|h:<b58-sha256>|d:<id>|t:<iso>Find a signature on any investigation page's decision log, or run python -m src.verify_decision --signature <sig> for a CLI check.
Decision
- Sequence
- #4
- Score
- 6 → 6 (0)
- Cluster
- mainnet-beta
- Slot
- 457444207
- Off-chain at
- 2026-10-16T00:52:49.264Z
- Anchored at
- 2026-10-16T00:52:51.709Z
- Block time
- —
Independent verification
- 1. Database (off-chain)
- 6Zh2uBphd7ixmZmPDpTgHHqMRUXvHjnS3RadeBn1YrHB
- 2. Recomputed (your browser)
- computing…
- 3. On-chain (Solana memo)
- fetching…
Canonical bytes hashed (2061 chars)
{"actor":"reviewer","artifact_identity":"6f08bbd041a2cb069dee6201df945c44","decided_at":"2026-10-16T00:52:49.017Z","decision":"review","disputed_pct":0.3913,"findings_hash":"BsN72sveMfyuLTwZrgaCwCiLBwx4ngNHaN27kMt8FwET","investigation_id":"cbe21877-b631-47a7-a344-fe90e23b96f0","mode":"findings_only","new_score":6,"page_content_hash":"6e7dd8dd0ae98a075ef70210e156ca86e8b7aff08708379d3a7cfea2798e6d53","page_slug":"bankr-bankrbot-ai-agent-prompt-injection-exploit","prev_score":6,"reason":"The page's core narrative — a two-stage NFT-privilege-escalation-plus-Morse-code prompt injection draining ~3B DRB tokens from Bankr's Grok-controlled wallet on May 4, 2026, followed by a second permission-chain breach affecting 14 wallets around May 19-20, 2026 — is well supported by independent reporting (CryptoTimes, Bitcoinist, SlowMist's secondary coverage, Giskard, OECD) and by on-chain data (Basescan confirms the cited wallet address and a matching outgoing transaction dated May 4, 2026). Most direct quotes (BankrBot, Grok, Yu Xian, OECD, CryptoDaily) check out verbatim or near-verbatim. Two categories of defect stand out: (1) a one-day date error affecting both prose and timeline fields for the second breach's reporting/disclosure sequence, which multiple outlets and Bankr's own dated X post place on May 18/19 rather than the page's May 19/20; and (2) a disputed market-impact figure (15-20% price decline) that conflicts with the more frequently reported ~40% figure, including from a source cited in the very same section. One attributed quote (to 'independent researcher Vadim') could not be corroborated anywhere, including in its own cited sources, and is flagged unverifiable rather than confirmed. A framing claim about the DRB token's origin ('unrelated user query') overstates the randomness of Grok's involvement — reporting shows a user specifically asked Grok to name a token for Bankr deployment.","score_delta":0,"sequence_num":4,"submission_content_hash":null,"submission_id":null,"submission_kind":null,"submission_valence":null,"v":1}