fix(engine): isolate getTenantConfig reads and normalize tenant config

[kai392]

Closes #9614

What

packages/loopover-engine/src/tenant-config.ts promised full per-tenant isolation but had two gaps:

1. The read path leaked the stored object. getTenantConfig returned store[tenantId] ?? resolveTenantConfig() — only the miss path was a fresh copy. On a hit it handed back the stored object, and setTenantConfig's Object.freeze is shallow (nested preferences/allowedActionClasses weren't frozen), so a caller could mutate a read config and silently rewrite that tenant's stored autonomyLevel/allowedActionClasses.

Fix: getTenantConfig now returns a deep copy on both paths, and setTenantConfig deep-freezes the written entry (config, its preferences, and the action-class array).

2. maxConcurrentLoops was unnormalized. ?? let -5, 0.5, NaN, Infinity pass through verbatim. It now uses the exact finiteNonNegativeInt body from tenant-quota.ts:45-47 (the #5828 discipline shared by loop-consumption.ts/worktree-pool.ts), applied to an explicitly-supplied override only — an absent override still inherits DEFAULT_TENANT_CONFIG's 1. resolveTenantConfig also now drops non-string allowedActionClasses entries rather than copying them through, mirroring the existing autonomyLevel guard.

DEFAULT_TENANT_CONFIG, TENANT_AUTONOMY_LEVELS, EMPTY_TENANT_CONFIG_STORE, and every exported type are unchanged; getTenantConfig's JSDoc is corrected.

Tests

5 new cases in test/unit/tenant-config.test.ts: maxConcurrentLoops normalization (-5/0.5/NaN/Infinity→0, 3→3, absent→1), non-string action-class dropping, deep-copy-on-hit isolation, and deep-freeze of the stored entry. 100% branch + statement coverage on the changed code; verified non-vacuous (4/5 fail with the fix reverted). tsc --noEmit clean.

[loopover-orb]

Caution

🛑 LoopOver review result - fixes required

_{Review updated: 2026-07-29 00:50:59 UTC}

2 files · 1 AI reviewer · no blockers · CI failing · unstable

🛑 Suggested Action - Fix Blockers

Review summary
The PR fixes two issues in tenant config handling: 1) `getTenantConfig` now returns deep copies on both hit/miss paths to prevent config mutation, and 2) `maxConcurrentLoops` is normalized to finite non-negative integers. The deep-copy logic uses a dedicated `cloneTenantConfig` function, and `setTenantConfig` deep-freezes all levels of the config object. The tests validate these changes with 100% coverage.

Nits — 2 non-blocking

• The `finiteNonNegativeInt` helper could benefit from a JSDoc comment explaining its purpose and behavior for clarity.
• Consider adding a test case for `allowedActionClasses` with mixed types (e.g., strings and numbers) to ensure non-string entries are consistently filtered.

CI checks failing

• validate
• changes

Decision drivers

• ✅ Code review — No blockers (1 reviewer)
• ✅ Gate result — Passing (No configured blocker found.)

Context & advisory signals — never blocks the verdict

Signal	Result	Evidence
Linked issue	✅ Linked	#9614
Related work	✅ No active overlap found	No same-issue or scoped active PR overlap found.
Change scope	✅ 20/20	Low review scope from cached public metadata (1 linked issue).
Validation posture	❌ 5/25	Preflight is holding this PR: the review lane is unavailable, so it is not ready for automated review.
Contributor workload	✅ 10/10	Author activity: 100 registered-repo PR(s), 49 merged, 7 issue(s).
Contributor context	✅ Confirmed Gittensor contributor	kai392; Gittensor profile; 100 PR(s), 7 issue(s).
Improvement	✅ Minor	risk: clean · value: minor

Linked issue satisfaction

Addressed
The PR implements deep copies for both paths in getTenantConfig via cloneTenantConfig, deep-freezes all nested structures in setTenantConfig, and normalizes maxConcurrentLoops with the required finiteNonNegativeInt logic. All required tests and JSDoc updates are included.

Review context

• Author: kai392
• Role context: outside_contributor
• Public audience mode: oss maintainer
• Lane context: Repository is registered but has no active allocation in the current snapshot.
• Public profile languages: not available
• Official Gittensor activity: 100 PR(s), 7 issue(s).
• PR-specific overlap: none found.

Contributor next steps

• Start here: Await review-lane availability.
• Then work through the remaining 2 steps in the Signals table above.

Signal definitions

• Related work = same linked issue, overlapping active PRs, or title/path similarity.
• Change scope = cached public metadata such as size labels, draft state, and review-burden hints.
• Validation posture = whether the PR provides enough public validation/test evidence for maintainer review.
• Contributor workload = public contributor activity and cleanup pressure, not a repo-wide quality failure.
• Contributor context = public GitHub/Gittensor identity context; non-Gittensor status is not a blocker.

🧪 Chat with LoopOver

Ask LoopOver a question about this PR directly in a comment — grounded only in the same cached, public-safe facts shown above, never a new claim.

• @loopover ask <question> answers contribution-quality Q&A with source citations and freshness.
• @loopover chat <question> answers in natural prose from cached decision-pack facts via local inference (maintainer/collaborator; read-only).
• A plain-language @loopover mention with a real question is routed to the closest matching read-only command automatically — no exact syntax required.

Full command reference: https://loopover.ai/docs/loopover-commands

🧪 Experimental — new and may change.

Decision record

• action: hold · clause: success
• config: b047b4811327bffb67b602f4aa4bb1b8cf82681e889f77a16f448399ec6add41 · pack: oss-anti-slop · ci: failed
• note: diverted by the randomized close-audit holdout (calibration: randomized ε-holdout on would-close PRs with propensity logging #8831) — the deterministic pipeline would otherwise have closed this PR
• record: 175b3caec5e4ee1babc33d259283b901c8f31af6f58f2fe89eae3d25cc983456 (schema v5, head 24dad48)

_{🟩 Safe / merged · 🟦 Advisory · 🟨 Held for review · 🟥 Blocked / closed}

---

💰 Earn for open-source contributions like this. Gittensor lets GitHub contributors earn for the work they already do — register to start earning →.

Checked by LoopOver, a quiet PR intelligence layer for OSS maintainers.

• Re-run LoopOver review

[superagent-security]

Superagent didn't find any vulnerabilities or security issues in this PR.

[loopover-orb]

LoopOver had already started reviewing this pull request — closing it to dodge the one-shot review process is not allowed. Please open a new pull request with the issues addressed.