Show HN: Traceseal – signed, offline-verifiable receipts for AI agent runs

Traceseal, a new open-source tool launched on Hacker News, provides signed, offline-verifiable receipts for AI agent runs, enabling any third party to verify agent actions without operator access. The receipts include skill identity, content-addressed manifests, input/output hashes, sandbox profiles, and operator signatures, and they chain into a transparency log. This comes as Article 50 of the EU AI Act, effective 2 August 2026, will require organizations to demonstrate what their AI systems did.

Verifiable receipts, ground-truth verification and replayable governance for agent work. A tamper-evident audit trail for every invocation — verifiable by anyone, no access to the operator required. Agents are doing real work — writing code, moving money, touching production. Traceseal is the accountability layer underneath: prove what ran, verify it against ground truth, replay it under governance. Every agent invocation emits a signed, self-contained receipt: skill identity, content-addressed manifest, input/output hashes, sandbox profile, operator signature. Attestation goes beyond "it ran": commands re-execute inside a kernel-namespace sandbox, and forbidden actions are enforced by the sandbox itself, not merely observed. Receipts chain into a transparency log. Any third party can re-verify the whole history with one command — no access to the operator's machine, keys or audit trail. Every receipt has three sections: what ran, who authorised the code, and who vouches for the execution. Any third party can verify all three. The publisher signs the skill bundle with an ed25519 key. The signature covers every source file via a content-addressed manifest, recorded in the transparency log. The operator runs the signed skill inside a kernel-namespace sandbox. The runtime records inputs, outputs, timing and sandbox configuration as hashes, then signs the record. Any third party installs traceseal-verify and runs one command. The verifier checks the operator's signature over the canonical JSON. No operator connection. No trust assumptions. Article 50 transparency obligations take effect on 2 August 2026. Organisations deploying AI systems will need to demonstrate what their systems did — not assert it. Signed execution receipts turn that from a documentation exercise into a one-command verification. Read Article 50 → https://artificialintelligenceact.eu/article/50/ Everything a third party needs to hold an agent to account is public — the receipt spec, the verifier, the runtime attestation tooling, and the transparency log itself. The independent verifier. One command, one receipt.json, zero trust assumptions. Ships with the open receipt spec. Runtime attestation: records execution inside the sandbox, hashes inputs and outputs, and signs the receipt with the operator key. Drop-in LangChain integration — signed receipts for chain and tool invocations without changing your agent code. The trust harness: reproducible behavioural probes that grade agent CLIs, emitting signed, witness-cosigned Traceseal receipts. Transparency log: log.traceseal.io https://log.traceseal.io No account. No API key. Just a receipt and the open-source verifier.