ENGINEERING
Block PRs that add incompatible Hugging Face model licenses
When a pull request adds or bumps a Hugging Face model dependency, it fetches the model card license, checks it against your org's allowed-license policy.
How it runs
The automated pipeline, trigger to output.
- TriggerPull request opened or updatedGitHub
- ActionExtract added HF model IDs from diffGitHub
- ActionFetch model card license for each modelHugging Face
- LogicCompare licenses against org allowlist
- OutputPost pass/fail GitHub status checkGitHub
What it does
Guards your repository against silently pulling in Hugging Face models whose licenses your legal policy forbids (for example a research-only or non-commercial license slipping into a commercial product). It runs on every pull request, identifies newly referenced model repos, resolves each one's license from its model card, and turns that into a required GitHub status check.
When to use it
Use it on any repo where engineers add `model_id` references, `from_pretrained` calls, or HF entries to a manifest, and where shipping a wrong-licensed model is a real legal risk. It is the enforcement layer that makes "check the license first" automatic instead of tribal knowledge.
How it works
- 1A GitHub pull request event fires the workflow.
- 2The diff is scanned to extract any added or changed Hugging Face model repo IDs.
- 3For each model ID, the Hugging Face API returns the model card metadata including the declared license tag.
- 4A logic step compares every license against your allowlist (for example apache-2.0, mit, bsd) and flags anything outside it.
- 5The workflow posts a GitHub commit status: green when all licenses pass, red with the offending model and license named when any fail, blocking merge until resolved.
Set it up
What you configure once, before turning it on.
- 1Connect GitHubRepos, issues, pull requests, actions.
- 2Connect Hugging FaceModels, datasets, spaces — the open-source hub.
- 3Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
- 4Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
- 5Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.
More Engineering workflows
Agent reviews model-license fit and suggests compliant swaps on the PR
When a PR adds a Hugging Face model, an agent reads the model card and license, judges fit against your commercial-use policy.
Quarterly Logging Hygiene Audit Agent
An agent-driven quarterly sweep that surveys all Axiom datasets, builds a logging-hygiene scorecard per service.
Post-Merge Log Volume Recheck After Downsampling PR
After a log-level PR merges, waits a day then re-queries Axiom to confirm the targeted stream's volume actually dropped.
Axiom Ingest Cost Spike to Linear Triage Ticket
When Axiom ingest volume spikes beyond its baseline, identifies which service caused it and files a Linear ticket with the offending log stream, sample lines, and a downsampling…
File a Linear license-review ticket for risky model adds
When a PR introduces a Hugging Face model with a non-permissive or unknown license, it opens a Linear issue assigned to the legal-review team with the model, license.
Warn the engineer in Slack when a model license is non-commercial
On a PR that adds a Hugging Face model, it checks the license and, if it is non-commercial or research-only, sends the PR author a direct Slack message explaining the restriction…
Run it inside a business
This workflow drops into a full company template. Import the org, and this is one of the playbooks its agents run.
Run this workflow in your colony.
14-day trial. No DevOps. No Sales call. Provisioned in under a minute.