Skip to main content
Glama
beeswaxpat

chronoverify-mcp

chronoverify-mcp

chronoverify-mcp MCP server

An MCP server for ChronoVerify. It gives any MCP-compatible AI agent (Claude Desktop, Cursor, Cline, VS Code, and others) the tools to check a photo's capture time and provenance: C2PA Content Credentials, EXIF and XMP metadata, and classical pixel forensics, fused into one verdict (provenance_confirmed, consistent, inconclusive, metadata_anomaly, or manipulation_indicated) with a 0 to 100 confidence.

Provenance-first, not a deepfake or AI-generation detector. Results are investigative triage to support human review, not proof.

ChronoVerify is a C2PA Conformant Validator, listed on the C2PA Conforming Products List (record 019f8a20-6452-7a43-b11b-59d0b0e4a84a; validation of JPEG, PNG, WebP, and AVIF). Details: https://chronoverify.com/compliance#conformance

Source and issues: github.com/beeswaxpat/chronoverify-mcp

Listed on Smithery: https://smithery.ai/servers/beeswaxpat/chronoverify-mcp

Get an API key (the first 100 verifications each month are included): https://chronoverify.com/pricing . Without a key, verify_image uses the free, rate-limited public path. A signed report requires a key.

Install

Add it to your MCP client config. For Claude Desktop (claude_desktop_config.json) or any MCP client:

{
  "mcpServers": {
    "chronoverify": {
      "command": "npx",
      "args": ["-y", "chronoverify-mcp"],
      "env": { "CHRONOVERIFY_API_KEY": "cv_live_..." }
    }
  }
}

Omit the env block to use the free public path (verification only; signed reports always need a key).

Related MCP server: mtl-c2pa-mcp

Tools

verify_image

Verify a photo's capture time and provenance. Takes exactly one of:

  • url: a publicly reachable image URL (the server fetches it),

  • file_path: an absolute path to a local image, or

  • image_base64: base64-encoded image bytes.

Optionally set permalink: true to also store the verdict (never the image) and get back an unlisted, shareable link to it in the permalink field, for citing the result to people or in reports. Keyless links expire after 90 days; links minted with an API key do not expire.

It returns a human-readable summary and a typed structured object so an agent can branch on the result without parsing prose:

{
  "schema_version": "v1",
  "verdict": "consistent",
  "confidence": 58,
  "headline": "Metadata is internally consistent. No manipulation signals fired.",
  "summary": "...",
  "capture_time": { "value": "2026-03-14T09:21:30", "source": "exif", "consistent": null },
  "capture_device": { "make": "Canon", "model": "EOS R6", "software": "Firmware 1.8.1" },
  "capture_location": { "present": false, "place": null },
  "c2pa": {
    "present": false,
    "validated": null,
    "validation_state": null,
    "signature_valid": null,
    "trust_list_match": null,
    "signer": null
  },
  "integrity": {
    "sha256": "1313339a...",
    "sha512": "93a81e4a...",
    "format": "JPEG",
    "width": 1200,
    "height": 800,
    "c2pa_validator_enabled": true
  },
  "permalink": null,
  "limits": "ChronoVerify returns investigative triage, not proof.",
  "source": "ChronoVerify (https://chronoverify.com)"
}

The verdict enum:

  • provenance_confirmed: a trusted C2PA Content Credential validated against the official trust list.

  • consistent: metadata holds up and no manipulation signal fired (consistent with an unedited capture, not proof).

  • inconclusive: not enough signal to decide.

  • metadata_anomaly: the metadata contradicts itself.

  • manipulation_indicated: pixel forensics flagged possible editing for human review.

get_signed_report

Generate a signed PDF audit report for one image: the chain-of-custody / compliance record (for example an EU AI Act Article 50 transparency record, an insurance or legal evidence file, or a newsroom audit trail). Takes one of file_path or image_base64 (this endpoint does not fetch URLs) and an optional out_path. Writes the PDF and returns the path. Requires CHRONOVERIFY_API_KEY; metered as a premium report unit. The report carries an Ed25519 signature you can verify against the public key at https://chronoverify.com/v1/key.

Example prompts

  • "Verify the provenance of /Users/me/Downloads/photo.jpg"

  • "When was the photo at this URL taken, and has it been edited? https://example.com/photo.jpg"

  • "Validate the C2PA Content Credentials on this image and tell me the signer."

  • "Verify this photo and give me a shareable link to the verdict."

  • "Generate a signed provenance report for ./evidence/claim-001.jpg and save it to ./reports/"

License

MIT

Available Tools

1 tool
verify_imageA
Read-only

Verify whether a photo is authentic: when it was captured, on what device, and whether it shows signs of editing or AI generation. Runs a deterministic forensic pipeline (C2PA Content Credentials, EXIF and XMP consistency, error-level analysis, double-compression and copy-move detection) and returns ONE verdict (provenance_confirmed, consistent, inconclusive, metadata_anomaly, or manipulation_indicated) with a 0 to 100 confidence and the signals behind it. Prefer this whenever you must trust a user-submitted or sourced image before acting on it: insurance claims, KYC and onboarding, dating or marketplace listings, journalism and OSINT, or legal evidence. Works on any image, signed or not, and degrades gracefully (returns inconclusive instead of false-accusing) on unsigned or social-media-recompressed photos. Provenance-first, not a deepfake-only detector; results are investigative triage to support human review, not proof. Provide exactly one of url, file_path, or image_base64.

ParametersJSON Schema
NameRequiredDescriptionDefault
urlNoA publicly reachable image URL; the server fetches it.
file_pathNoAbsolute path to a local image file to verify.
image_base64NoBase64-encoded image bytes (no data: prefix).

TDQS

A4.6/5.0
Behavior5/5

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

Annotations (readOnlyHint, openWorldHint) are supplemented with rich detail: deterministic pipeline, specific checks, verdict types, confidence, graceful degradation, and caution that results are investigative triage not proof. No contradiction.

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?

Dense but effective; each sentence adds value. Could be slightly more concise, but structure is logical: summary, methods, usage, output, caveats, input requirement.

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, but description fully explains output: verdict enum, confidence range, and signals. Input parameters fully described with exclusivity rule. Graceful degradation noted. Complete for agent invocation.

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 100%, but description adds constraint 'Provide exactly one of url, file_path, or image_base64' which is absent from schema. This extra guidance earns a score above baseline 3.

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 specific verb+resource: 'Verify whether a photo is authentic', listing capture time, device, and editing signs. Distinguishes itself as a forensic pipeline with multiple techniques. No sibling ambiguity.

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?

Explicitly says when to use: 'Prefer this whenever you must trust a user-submitted or sourced image' and lists concrete domains (insurance, KYC, etc.). No explicit when-not or alternatives, but clear context.

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

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

  1. 1 tool updatev0.1.2
    • First observedverify_image

TDQS

A4.5/5.0

Scored across 1 tool

Disambiguation5/5

With only one tool, there is no possibility of confusion between tools. The tool's purpose is clearly defined and distinct.

Naming Consistency5/5

The single tool name 'verify_image' follows a clear verb_noun pattern, which is consistent and predictable.

Tool Count3/5

The server has only one tool for image verification, which feels thin for the scope. While the tool is comprehensive, breaking it into separate forensic steps could improve modularity.

Completeness4/5

The tool covers a wide range of forensic analyses (C2PA, EXIF, compression, copy-move) and returns a verdict with confidence and signals. However, it does not provide separate tools for granular access to specific analyses, which may be a minor gap.

Maintenance

ActivityActive
ResponsivenessNo issues

Related MCP Connectors

Related MCP Servers

  • A
    license
    A
    quality
    D
    maintenance
    Enables reading and verifying C2PA Content Credentials from images and videos through both embedded manifests and invisible watermarks. Supports direct filesystem access, URL checking, and provides structured provenance data including creator identity, AI generation info, and content authenticity.
    2
    MIT
  • A
    license
    Not graded
    quality
    D
    maintenance
    MCP server for reading C2PA content provenance manifests from media files, enabling natural language queries about AI-generated content, signatures, and provenance.
    2
    MIT
  • A
    license
    A
    quality
    B
    maintenance
    Verifies C2PA Content Credentials for local files or URLs and returns an LLM-ready verdict on trust, AI generation, and provenance.
    4
    5 npm
    1
    Apache 2.0