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 · Coinkite / Coldcard
- Sequence
- #4
- Score
- 13 → 13 (0)
- Cluster
- mainnet-beta
- Slot
- 457579930
- Off-chain at
- 2026-10-16T09:01:27.705Z
- Anchored at
- 2026-10-16T09:01:38.327Z
- Block time
- —
Independent verification
- 1. Database (off-chain)
- 7TTZmarg6K4bFgUCLF2ta2K4wCYnBRE2XEwWyGNGge55
- 2. Recomputed (your browser)
- computing…
- 3. On-chain (Solana memo)
- fetching…
Canonical bytes hashed (2085 chars)
{"actor":"reviewer","artifact_identity":"1002d185934a6303d89dafc51c2a87a7","decided_at":"2026-10-16T09:01:27.484Z","decision":"review","disputed_pct":0.1304,"findings_hash":"AXGte2cW6y1Hf43svK6y2zqJqoDooYEU5KRXzFePfMZe","investigation_id":"fa3e4739-2493-471f-b16a-4a0503f66079","mode":"findings_only","new_score":13,"page_content_hash":"d57a2db786d5e008b360c5d1f27819c95230a56caee4a25f154ca5d19df40a48","page_slug":"coinkite-coldcard","prev_score":13,"reason":"The page's core narrative — the March 2021 libngu/Yasmarang RNG regression in Coldcard firmware 4.0.0, its July 2026 exploitation across multiple waves totaling roughly $116-130M from 5,200+ addresses, Coinkite's emergency patches and data-retention suspension, the O'Beirne prior-warning account, and the Peter Gray/'switck' GPG-signature allegation — is well-supported by primary and credible secondary sources, including Coinkite's own firmware changelog and the TRM Labs report. Figures that vary across outlets (Fortune's $116M vs. CBC/Galaxy's up to $155M) are accurately presented as time-stamped snapshots of an evolving investigation rather than a single disputed number. The most substantive problems found: (1) the section heading's 'CVE:' label implies a formal CVE identifier that no source confirms exists; (2) the claim that North Korea was 'specifically noted as unlikely' overstates what TRM Labs actually said (a laundering-pattern contrast, not an affirmative exculpation); (3) a quote attributed to Kraken's CSO is partly the source article's paraphrase rather than a verbatim statement; (4) one cited source (Fieldfisher) is inaccessible both live and via archive, leaving an unverifiable quote about recovery obstacles; and (5) the 40/72-bit entropy figures are presented as settled when independent technical sources differ. None of these rise to fabrication; they are precision and attribution issues layered on an otherwise solid, well-hedged investigation.","score_delta":0,"sequence_num":4,"submission_content_hash":null,"submission_id":null,"submission_kind":null,"submission_valence":null,"v":1}