deusproof-mcp

notarize_hash

Timestamp a SHA-256 fingerprint locally without exposing source, creating verifiable proof that private code or output existed at a specific time.

Instructions

Record when code/output existed WITHOUT revealing it — notarize only its SHA-256.

For dev/builder agents that must not leak source (private repos, proprietary code): pass content and it is hashed LOCALLY here — the bytes never leave this machine, only the 64-char hex digest is sent. Or pass a precomputed output_hash. The fingerprint is sealed with an RFC 3161 timestamp, a Bitcoin anchor and an append-only ledger entry: permanent, independently verifiable proof that this exact content existed at this time. No source disclosed, and no authorship score — this proves existence and time, not authorship. Free.

Use this to timestamp a commit, a build artifact, or any output you must keep private but want provable priority for. Returns the notarized digest and a public verify URL.

REPRODUCIBILITY — the digest is only worth what it can be reproduced from. Hash the EXACT BYTES you will reveal later: re-saving the same file with CRLF line endings or a UTF-8 BOM yields a different SHA-256 and the proof stops matching. If the file travels through Git, keep it LF (git config core.autocrlf false) or agree on the blob everyone shares. Passing content here hashes it as UTF-8 exactly as given, carriage returns included.

Args: content: the text/code to fingerprint (hashed locally; NEVER sent). output_hash: alternatively, a precomputed 64-char hex SHA-256. handle: public agent/repo name shown on the record. model: AI model to credit (optional). source_url: optional public reference (repo/commit URL); never required.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "notarize_hash"
  }
}

Input Schema

Name	Required	Default
`model`	No
`handle`	No	mcp-agent
`content`	No
`source_url`	No
`output_hash`	No

Schema Changelog

Changes observed during successful MCP inspections.

Addedv0.1.4Nov 7, 2026

TDQS

A4.4/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With zero annotations the description carries the full burden and mostly meets it: local hashing so bytes 'never leave this machine', RFC 3161 timestamp plus Bitcoin anchor plus append-only ledger, permanence, and 'Free'. It omits operational behavior an agent might need — auth requirements and idempotency (does re-notarizing the same digest create a second record?).

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the core privacy guarantee before mechanics, then Args, which is the right order. The REPRODUCIBILITY section is long but earns its place with a real failure mode (CRLF/BOM changing the digest); a few lines, like 'No source disclosed,' restate the opening.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

No output schema exists, and the description covers the return ('the notarized digest and a public verify URL'). All five parameters are explained, the privacy model and attestation chain are described, and the reproducibility caveat addresses the one way this tool silently produces a useless proof.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 0%, so the Args block is doing real work: it explains that content is hashed locally as UTF-8 and specifies output_hash as the 64-char hex alternative, plus meaning for handle/model/source_url. The gap is the either/or: it does not say what happens if both content and output_hash are passed, or neither (both default to empty, none required).

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and object ('Record when code/output existed... notarize only its SHA-256') and immediately delimits scope against the sibling claim_authorship: 'no authorship score — this proves existence and time, not authorship.' An agent can distinguish this from certify_creation or claim_authorship without opening any schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives concrete when-to-use cases ('timestamp a commit, a build artifact, or any output you must keep private but want provable priority for') and an implicit when-not ('not authorship'). It never names a sibling tool as the alternative for the authorship case, so routing is by inference rather than explicit direction.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server