Per-slot allowlist scoping for safety-checker slots
@JAORMX is already working on this.
Since Aug 12, 2026.
- Dominant language
- Go
- Stars
- 152
- Forks
- 16
- Avg merge
- 14h 48m
- Merged PRs (30d)
- 536
Description
Documented follow-up from PR #87 (https://github.com/stacklok/mecatl/pull/87) — not in scope for the model-selection slice.
Current state
The operator allowlist (Claude Code's availableModels shape) caps which model ids a trusted project may bind to models.default/slots/aliases. But an allowlisted model may currently be bound to any slot, including the safety checkers (the compaction summary, ask-reviewer, and guardrail checker slots). Today this is covered only by documented operator guidance.
Work
Add per-slot allowlist scoping so an operator can constrain which models are eligible for the safety-critical slots specifically (tighten-only, operator-tier). Mind the existing invariants:
- The allowlist cap is operator-tier only (project-tier ignored without an allowlist ⇒ opt-in).
canonicalAllowlistresolves operator entries BEFORE project aliases merge;TestProjectAliasLaunderingDroppedverifies a project alias can't launder an unvetted id.- Trust gate enforced at both layers.
See ADR 0030 §Phase 4.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.