AI-BOM License Chain Resolver for Hugging Face Models

$0.50 USDC per call · Developer and security · POST /api/v1/ai-bom-license-resolver · Id ai-bom-license-resolver

Walks the Hugging Face lineage of up to 10 models (base_model chain, training datasets; at most 30 nodes, 4 generations) and applies a curated, hash-pinned table of license clauses (Apache-2.0, MIT, CC-BY, CC-BY-NC, CC-BY-SA, OpenRAIL variants, Llama 2 / 3 / 3.1 / 3.2 / 3.3, Gemma, Mistral Research License) to each intended use: commercial or not, API, weights or on-prem, fine-tuning, monthly active users, jurisdiction, sensitive fields of use. Per use it returns ALLOWED, CONFLICT or UNDETERMINED with the conflicting nodes, the clause, a verbatim quote, and the license source URL and SHA-256; the unsatisfiable core comes from the hub's deterministic SAT/LRA solver, with no LLM. Unknown, custom, missing or unreadable licenses are UNDETERMINED, never ALLOWED. Not legal advice.

What you get

Tells you whether a model stack may be used for your stated use, names the conflicting model and clause, and ships a CycloneDX-style AI-BOM.

Call it

Send the JSON body below to https://agents.writerslogic.com/api/v1/ai-bom-license-resolver. An unpaid request answers 402 with the price and how to pay; a request that fails with a 5xx status is not charged.

curl -i -X POST https://agents.writerslogic.com/api/v1/ai-bom-license-resolver \
  -H 'content-type: application/json' \
  -d '{"models":["NousResearch/Hermes-3-Llama-3.1-8B","mistralai/Mistral-7B-Instruct-v0.2"],"uses":[{"id":"api-above-700m-mau","commercial":true,"distribution":"api","monthly_active_users":800000000},{"id":"api-below-700m-mau","commercial":true,"distribution":"api","monthly_active_users":50000000}]}'

Input

Field	Type	Required	Description
`models`	array<string; length 1..200>; max 10 items	yes	Hugging Face model repo ids (owner/name) used together in the system.
`datasets`	array<string; length 1..200>; max 10 items	no	Extra Hugging Face dataset repo ids; datasets named on the model cards are followed automatically.
`uses`	array<object>; max 6 items	yes	Intended uses to decide, each evaluated independently.
`dataset_terms`	string; one of "advisory", "binding"	no	advisory (default): a dataset license restriction yields UNDETERMINED because it is unsettled whether it binds trained weights. binding: treat it as binding and report CONFLICT.

Example request body

{
  "models": [
    "NousResearch/Hermes-3-Llama-3.1-8B",
    "mistralai/Mistral-7B-Instruct-v0.2"
  ],
  "uses": [
    {
      "id": "api-above-700m-mau",
      "commercial": true,
      "distribution": "api",
      "monthly_active_users": 800000000
    },
    {
      "id": "api-below-700m-mau",
      "commercial": true,
      "distribution": "api",
      "monthly_active_users": 50000000
    }
  ]
}

Result

The response is { "data": ..., "_provenance": ... }. data carries these fields; _provenance is a signed W3C Verifiable Credential with a C2PA-style manifest, checked free at POST /api/v1/verify-provenance.

Field	Type	Required	Description
`verdict`	string; one of "ALLOWED", "CONFLICT", "UNDETERMINED"	no	Worst per-use verdict: CONFLICT, then UNDETERMINED, then ALLOWED. Unknown, custom, or missing licenses are never ALLOWED.
`confidence`	string; one of "high", "low"	no	Completeness of the rule evaluation, not legal certainty: low when any use is UNDETERMINED.
`deterministic`	boolean	no	Always true: no LLM touches the verdict path.
`evidence`	array<object>	no
`summary`	string	no
`next_step`	string	no
`not_legal_advice`	string	no
`uses`	array<object>	no
`missing_license_nodes`	array<object>	no
`license_mismatches`	array<object>	no
`lineage`	object	no	nodes (key, id, kind, depth, license status and id, source_url), edges (from, to, relation), truncations, limits.
`ai_bom`	object	no	CycloneDX 1.6 style ML-BOM: components, licenses, dependencies.
`license_table`	object	no	Table version, verification date, and for every license used: sources with URL and SHA-256.
`source_manifest`	array<object>	no
`partial`	object	no	complete flag and the list of what is missing when an upstream read failed or a limit cut the lineage.
`solver`	object	no	Solver name, version, and total solver calls.

Pay and connect

Available through: x402 (USDC on Base), MCP, RapidAPI, Apify.

Machine-readable

This agent as JSON (schemas, example, OpenAPI operation)
Free signed sample response
OpenAPI 3.1 · Catalog · Index of every agent