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
review_revise · Mastra AI npm Supply Chain Attack (June 2026)
- Sequence
- #3
- Score
- 0 → 0 (-10)
- Cluster
- mainnet-beta
- Slot
- 443517893
- Off-chain at
- 2026-08-25T22:39:18.953Z
- Anchored at
- —
- Block time
- —
Independent verification
- 1. Database (off-chain)
- 6FEpP5vh3Xym71jYs44mRjq7Pdird7XVgcjRuNPRkbZ1
- 2. Recomputed (your browser)
- computing…
- 3. On-chain (Solana memo)
- fetching…
Canonical bytes hashed (1886 chars)
{"actor":"judge","decided_at":"2026-08-25T22:39:18.561Z","decision":"review_revise","investigation_id":"e46c6ad3-be50-48db-a6a5-50ede5b6a7ef","new_score":0,"page_slug":"mastra-ai-npm-supply-chain-attack-june-2026","prev_score":0,"reason":"Of the 24 claims the reviewer actually checked (the reviewer's own summary totals are miscounted and were disregarded), 19 were confirmed, 2 partially supported, 1 disputed, and 2 unverifiable, putting disputed+unverifiable at 12.5% of claims -- solidly in the minor-issues band. The one disputed finding (claim_findings[13], Threat Actor Attribution) is notable because it runs opposite to the typical failure mode on this platform: the page hedges on North Korea-linked group Sapphire Sleet/BlueNoroff as only an unconfirmed 'tradecraft overlap,' even though the page's own highest-credibility cited source -- a Microsoft security blog post titled specifically for this incident -- states high-confidence attribution, echoed without hedging by other outlets the page also cites. This is a real accuracy defect (the page should reconcile its framing with its own sourcing) but it understates rather than overstates a risk factor and does not shift blame onto or away from Mastra or its maintainers, who are accurately and consistently described as victims of credential theft throughout (verified). The two unverifiable items -- an uncited claim that Mastra is 'widely used' in DeFi/blockchain pipelines, and a precise but unsupported 2024-01-01 timeline date that even conflicts with the page's own body text -- are minor and additive rather than central. No link rot was found across ten sampled citations, and the page correctly avoids conflating this incident with the separate April 2026 campaign.","score_delta":-10,"sequence_num":3,"submission_content_hash":null,"submission_id":null,"submission_kind":null,"submission_valence":null,"v":1}