bernstein-mcp
Server Details
Verifies Bernstein run receipts and hash chains; lists the shipped presets and adapters. Read-only.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- Repository
- sipyourdrink-ltd/bernstein-mcp
- GitHub Stars
- 0
- Server Listing
- bernstein-mcp
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 12 tools
Most tools target clearly different artifacts (receipt, chain rows, delegation chain, agent manifest, presets, adapters), but `trace_verify_record` and `verify_trace_record` both appear to verify a single TRACE v0.2 Trust Record and an agent would struggle to pick between them without reading the full descriptions. `verify_chain` and `verify_delegation_chain` also sit close together, though the description's row-kind detection and delegation walk differentiate them. The explain_* vs verify_* split on receipts is otherwise clean.
Ten of twelve tools follow a verb_noun snake_case pattern (verify_receipt, verify_chain, list_presets, get_preset), which is readable and predictable. However `server_info` drops the verb entirely and `trace_verify_record` inverts the order to noun_verb while a near-identical `verify_trace_record` also exists — a deviation that actively worsens confusion rather than just being stylistic.
Twelve tools is squarely in the well-scoped range for a verification/attestation server, and each tool maps to a distinct artifact family or lookup (receipts, chains, TRACE records, delegation chains, manifests, presets, adapters, server metadata). No filler tools are apparent beyond the redundant TRACE pair.
The verification surface is broad: single-record and chain verification for receipts, journal/lineage rows, TRACE records and delegation chains, plus manifest conformance and preset/adapter introspection. Gaps remain on the discovery/emission side — there is no way to fetch or list runs by identifier, no tool to emit or assemble a TRACE record (explain_trace_mapping only maps), and no preset/policy linting — but agents can work around these by supplying artifacts directly.
Available Tools
12 toolsexplain_receiptExplain a run receiptBInspect
Verify a run receipt and narrate the result in plain language: what the run recorded, which chains recomputed, where the first divergence is, and what an auditor can and cannot conclude without the producing install's keys.
Inputs: receipt (required). Returns: verdict, explanation, receipt_sha256
explain_trace_mappingExplain the bernstein → TRACE mappingAInspect
How a bernstein run maps onto a TRACE v0.2 Trust Record: one row per claim with the journal field it is sourced from and the rule that derives it. Pass a run receipt to fill in the rows its embedded journal can answer (subject, iat, model, data_class, policy digest, tool transcript) alongside the receipt's own verdict.
Inputs: receipt. Returns: mapping, verdict, markdown, receipt_sha256
get_presetGet a compliance presetBInspect
Every field of one compliance preset as bernstein v3.19.2 resolves it.
Inputs: name (required). Returns: name, config, bernstein_version
list_adaptersList agent adaptersAInspect
The agent adapters bundled with bernstein v3.19.2: adapter name, the binary it drives, the module that implements it.
Inputs: no named parameters. Returns: adapters, bernstein_version
list_presetsList compliance presetsAInspect
The compliance presets bernstein v3.19.2 ships, with the switches each one turns on.
Inputs: no named parameters. Returns: presets, bernstein_version
server_infoServer infoAInspect
Identity, version, and request limits for this MCP server.
Inputs: no named parameters. Returns: name, limits, version, keys_url, +1 more
trace_verify_recordVerify a TRACE record with trust-record-verifyAInspect
Verify a TRACE v0.2 Ed25519 Trust Record locally with optional caller public key and freshness controls. Nothing is fetched. Embedded keys establish internal consistency only; revocation is not affirmed.
Inputs: record (required), now, trusted_key, max_age_seconds. Returns: note, message, profile, verdict, +5 more
verify_agent_manifestVerify an Agent ManifestAInspect
Agent Manifest v0.2 conformance checks on one manifest, stateless, no account: schema (vendored agent-manifest.schema.json), profile context, version, canonicalization, COSE envelope signature (Ed25519, key identified by kid in the protected header; ML-DSA-65 is reported unverifiable). Optionally checks that a TRACE Trust Record cites this manifest by digest. Nothing is fetched; resolvers are checked as URIs only. The manifest hash is sha256 over the COSE payload bytes (or RFC 8785 canonical JSON for object input).
Inputs: manifest (required), trustRecord. Returns: note, checks, summary, verdict, +2 more
verify_chainVerify a hash chainAInspect
Walk bernstein chain rows and recompute every link: journal rows (event_hash), lineage spine entries (entry_hash) or audit events (prev_hmac linkage). The row kind is detected from the fields, or pass kind explicitly. Reports the first divergent index. Pass entries as the file text (a JSON array or one row per line, as journal.jsonl is written) for byte-exact hashing; a parsed array also works, but then a float spelled 1.0 or -0.0 in the file cannot be told apart from an integer.
Inputs: entries (required), kind. Returns: head, kind, detail, intact, +2 more
verify_delegation_chainVerify a TRACE delegation chainAInspect
TRACE v0.2 delegation-chain conformance, stateless, no account: index every Trust Record by the RFC 8785 digest of its complete form, start at the leaf, follow delegation.parent_record_hash to the root, and check each hop's signature, the root key against trusted_root_keys, the depth bound, the link's digest algorithm, the credential (registered, issuer = parent subject, holder = record subject, window at the hop's own iat) and data_class narrowing under data_class_lattice. Classification: provenance-invalid outranks authorization-invalid; an unread link is unverifiable, not broken. Pass records as a JSON array or as file text (one record per line or a JSON array).
Inputs: records (required), context. Returns: note, walk, codes, depth, +4 more
verify_receiptVerify a run receiptAInspect
Recompute every hash chain a bernstein run receipt embeds (journal, lineage spine, optional audit range), rebuild the signed subject from the recomputed heads, and check the Ed25519 signature with the key the receipt carries. Needs no secret and reads nothing but the receipt. Returns the verdict, the first failing check, one line per check, and the same verdict as a DSSE envelope signed by this verifier's Ed25519 key (public key at keys_url) so the outcome can be kept and re-checked offline.
Inputs: receipt (required). Returns: note, checks, summary, verdict, +6 more
verify_trace_recordVerify a TRACE Trust RecordAInspect
TRACE v0.2 conformance checks on one Trust Record, stateless, no account: schema (vendored trace-claim.json), profile, subject URI, software-only runtime rule, policy digest, public-only confirmation key, the embedded signature (EdDSA, ES256 or ES384 with the key in cnf.jwk), appraisal, delegation link shape and references. Nothing is fetched: resolvers are checked as URIs only. record_sha256 is the RFC 8785 digest of the complete record, signature included — the value a child hop puts in delegation.parent_record_hash.
Inputs: record (required). Returns: note, checks, summary, verdict, +2 more
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Read-only verifier for 25 ProofRelay MCP tools and non-confidential evidence bundles.
Find governed AI capabilities and verify signed receipts. Read-only, no account.
Issue & verify signed (ed25519), hash-chained, timestamped provenance receipts for agent actions.
Read-only gateway for durable agent identity, consent, recognized work, and signed receipts.
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceEnables verifying signed provenance receipts by ID or full receipt object and listing an agent's receipt hash-chain newest first for auditing.MIT
- AlicenseNot gradedqualityCmaintenanceEnables users to run exposure audits, promotion gates, bank-health checks, counterexample hunts, and other verifier-backed tools on disposable data, returning inspectable certificates, SHA-256 receipts, and downloadable artifacts.AGPL 3.0

cwi-mcp-serverofficial
AlicenseAqualityBmaintenanceProvides read-only access to CWI's trust infrastructure, enabling Gear Ledger provenance queries, agent trust scoring via the Verdict Engine, and NEEDLE DROP ledger hash-chain verification over stdio.7MIT- FlicenseNot gradedqualityDmaintenanceEnables read-only verification of Foster Rx certificates against the public Ed25519 trust anchor, returning verdicts such as verified, signature_invalid, not_found, or tool_fault, and also allows retrieval of certificate records and trust anchor material.-
Glama MCP Gateway
Add one secure layer between your agents and this server.