feat(api): add installation-scoped bulk agent pause/dry-run endpoint

[JSONbored]

Summary

• Adds PUT /v1/app/installations/:id/agent/bulk-settings, applying agentPaused/agentDryRun across every currently-installed repo in a tenant's installation in one call — previously only strictly per-repo via PUT /v1/repos/:owner/:repo/settings.
• Reuses this route family's existing tenant-scoping exactly (resolveAppInstallationScope/installationRecordInScope): an operator can write any installation; a non-operator session only their own — enforced before any write.
• Distinct from the global operator kill-switch (getGlobalAgentFrozenState), untouched by this.
• Own audit-trail event (installation.agent_bulk_settings_updated), distinct from both per-repo edits (repo.settings_updated) and the kill-switch's own event.

Closes #7676.

Test plan

• New tests in test/integration/app-installations-selfservice.test.ts: auth/role rejection, invalid installation id / unknown installation / malformed body (incl. genuinely unparseable JSON) / unrecognized field, tenant-isolation (never lets tenant A touch tenant B's installation, enforced pre-write), applies across every installed repo and only those repos, partial-field updates preserve the other flag, no-op for zero-repo installations, operator bypass with its own audit event.
• npx tsc --noEmit
• Targeted coverage check — 100% line/branch/function coverage on every new line
• npm run ui:openapi:check / npm run ui:openapi:settings-parity — clean
• npm run test:coverage (unsharded, full suite) — clean
• npm audit --audit-level=moderate — 0 vulnerabilities
• npm run engine-parity:drift-check / npm run manifest:drift-check — clean

[superagent-security]

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

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 90.11%. Comparing base (4578fef) to head (896de6a).
✅ All tests successful. No failed tests found.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #8144      +/-   ##
==========================================
- Coverage   92.03%   90.11%   -1.92%     
==========================================
  Files         764       98     -666     
  Lines       77520    25151   -52369     
  Branches    23430     4702   -18728     
==========================================
- Hits        71346    22665   -48681     
+ Misses       5062     2214    -2848     
+ Partials     1112      272     -840     

Flag	Coverage Δ
shard-1	29.85% <100.00%> (-27.81%)	⬇️
shard-2	43.32% <10.00%> (-7.57%)	⬇️
shard-3	70.72% <10.00%> (+16.25%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
src/api/routes.ts	95.27% <100.00%> (+0.03%)	⬆️

... and 666 files with indirect coverage changes

[loopover-orb]

Tip

✅ LoopOver review result - approve/merge recommended

_{Review updated: 2026-07-23 04:53:11 UTC}

2 files · 1 AI reviewer · no blockers · readiness 100/100 · CI green · clean

✅ Suggested Action - Approve/Merge

• safe to merge

Review summary
This adds `PUT /v1/app/installations/:id/agent/bulk-settings`, reusing the existing `resolveAppInstallationScope`/`installationRecordInScope` tenant-isolation pattern from the health/repair routes and applying `agentPaused`/`agentDryRun` across every installed repo in an installation via a `.strict()` picked-field schema. The forbidden-installation check runs before body parsing and before any write, and partial-field updates correctly preserve the other flag by merging into `getRepositorySettings` first. Test coverage is thorough (auth, scope, malformed/unparseable JSON, strict-extra-field rejection, partial update, zero-repo no-op, operator bypass with distinct audit event) and closes the linked #7676.

Nits — 6 non-blocking

• src/api/routes.ts: the `Promise.all` over `repoFullNames.map` fires one `getRepositorySettings` + `upsertRepositorySettings` pair concurrently per installed repo with no batching/limit — fine for typical tenant sizes but worth a comment or cap if an installation can have hundreds of repos.
• src/api/routes.ts: an empty body `{}` passes `.safeParse` (all fields optional) and still performs a full write pass across every repo plus a `recordAuditEvent` call for a functionally no-op change — consider short-circuiting when `Object.keys(changes).length === 0`.
• The four inline HTTP status literals (400/403/404) flagged by tooling match this file's existing per-route convention (see the sibling health/repair routes), so no change needed there.
• Consider adding a quick `Object.keys(changes).length === 0` early-return to skip the repo fan-out and audit write for a genuinely empty body.
• If installations can have very large repo counts, consider chunking the `Promise.all` to bound concurrent D1 writes.
• PR author also opened the linked issue — Link an issue that was opened by a different contributor, or provide a rationale for why this self-authored issue represents genuine discovery work.

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	#7676
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	✅ 25/25	PR body includes validation/test evidence.
Contributor workload	✅ 10/10	Author activity: 14 registered-repo PR(s), 14 merged, 228 issue(s).
Contributor context	✅ Confirmed Gittensor contributor	JSONbored; Gittensor profile; 14 PR(s), 228 issue(s).
Improvement	✅ Minor	risk: clean · value: minor · LLM: significant

Linked issue satisfaction

Addressed
The PR adds a new installation-scoped PUT /v1/app/installations/:id/agent/bulk-settings endpoint that applies agentPaused/agentDryRun across all repos in an installation with proper tenant-scoping, and records a distinct audit event (installation.agent_bulk_settings_updated) separate from per-repo edits and the global kill-switch, matching all stated deliverables and boundaries.

Review context

• Author: JSONbored
• Role context: owner (maintainer lane)
• Public audience mode: oss maintainer
• Lane context: Repository is configured for direct PR review.
• Public profile languages: Python, TypeScript, Ruby, Go, JavaScript, MDX, Shell, Solidity
• Official Gittensor activity: 14 PR(s), 228 issue(s).
• PR-specific overlap: none found.

Contributor next steps

• Start here: Treat this as maintainer-lane context rather than normal contributor-lane activity.

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.

_{🟩 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

[loopover-orb]

LoopOver approves — the gate is satisfied and CI is green.