Official
Regulated default
- Published and signed by ElixirHub
- Architecture review
- Full evaluator suite
- 12-month deprecation notice
Agent skills supply chain · Part of the Elixir Platform
ElixirHub scans, tests, signs, and version-pins agent skills before they reach a workspace. Follow each version to the decisions it shaped, so auditors can trace it end to end.
See how a skill reaches your workspaceThe problem
Skills bring new instructions and sometimes executable code. Before a skill reaches production, teams need to know what it can access, which version will run, and who approved it.
Why skills need a supply chainHow it works
The publisher builds the skill and declares its tools and inputs in the manifest.
Review the package and run sandbox, functional, adversarial, and regression tests.
Bind the publisher key and attestations to the release’s content hash.
Make the immutable version discoverable by trust tier and domain.
Check that the release’s tier meets workspace policy.
Skill testing
Evaluate each version before signing. Repeat the tests when models, tools, or test cases change to catch regressions before rollout.
Run scripts in isolation and flag access the skill did not declare.
Score each test case against expected results and declared failure modes.
Check whether misleading inputs cause a skill to expose data or exceed its permissions.
Evaluate private workspace cases and catch regressions before rollout.
Why ElixirHub
See who approved the release, what it was tested against, and where it influenced a decision.
Part of the Elixir Platform
Three products cover the registry, the agent’s actions, and the record of what was decided.
Publish, scan, evaluate, sign, and version skills before teams install them.
Apply permissions, decision boundaries, and guardrails at the Agent Gateway.
Use the Context Graph to trace decisions back to the skill version that informed them.
Trust tiers
Set which skill releases a workspace can install based on who published them and what evidence they carry.
Regulated default
Opt in
Off by default
Compare
Scroll to see the full comparison.
| Capability | Typical artifact registry | ElixirHub |
|---|---|---|
| Version, signing, and access control | Core registry functions | Pinned releases and workspace policies |
| Security scan before install | Package scanning | Scan evidence alongside release metadata |
| Functional, adversarial, and regression tests | External test tooling | Evaluation evidence by release |
| Domain golden sets | External evaluation tooling | Expected outcomes for regulated workflows |
| Decision traceability | Artifact metadata | Linked to the decision record in ElixirData |
Built for regulated work
Reconciliation, KYC review, SOX and IFRS controls, and materiality checks.
Explore skillsBilingual notices, case evidence, and licence checks.
Explore skillsKnowledge search, documentation quality, and operations workflows.
Explore skillsLog correlation, alert triage, and dependency analysis.
Explore skillsFor security and AI leaders
Map where skills come from, what runs in production, and which controls to put in place first.
Book an assessmentFor skill publishers
Prepare your skill and submit a version for review in the ElixirHub catalog.
Publish to ElixirHub