chore(models): refresh OpenCode fallback catalog

[simple-agent-manager]

Summary

• Refresh the OpenCode Zen and Go static fallback catalog from the live Models.dev provider records.
• Add six active model IDs and remove the inactive opencode/hy3-free fallback.
• Strengthen focused regression coverage for representative additions and the removed ID.

Primary sources consulted: Anthropic, OpenAI, Google, Mistral, and Models.dev.

Validation

• pnpm lint
• pnpm typecheck
• pnpm test
• pnpm build
• Focused shared catalog tests: 2 files, 21 tests passed
• pnpm quality:observability-noise reported no significant log noise (D1 ID absent; Workers telemetry returned 403 and those subqueries were skipped)
• Control-loop I/O budget: N/A; no sweep, cron, alarm, or candidate-selection logic changed

Staging Verification (REQUIRED for all code changes — merge-blocking)

• Staging deployment green — Deploy Staging run 30253600731 passed
• Live app verified via Playwright — authenticated with the designated staging token and exercised the live app
• Existing workflows confirmed working — dashboard, projects, settings, API tokens, auth redirect, API health, and CORS passed
• New feature/fix verified on staging — live catalog/settings smoke coverage passed, including Amp/catalog exposure; this change only alters the OpenCode fallback used when the dynamic Models.dev request is unavailable
• Infrastructure verification completed — N/A: no infrastructure paths changed
• Mobile and desktop verification notes added — N/A: no UI paths changed

Staging Verification Evidence

The authenticated Playwright smoke suite completed with 11 tests passing directly and one dashboard test passing on its configured retry after an initial 15-second networkidle timeout. No functional assertion failed after retry. The deployment workflow's own health and smoke jobs also passed.

UI Compliance Checklist (Required for UI changes)

N/A: no UI files changed.

End-to-End Verification (Required for multi-component changes)

N/A: single shared static fallback catalog plus its focused unit tests; no multi-component data flow changed.

Data Flow Trace

N/A: no multi-component feature.

Untested Gaps

The dynamic Models.dev success path is unchanged. The updated static outage fallback is covered by shared unit tests and live catalog smoke coverage.

Post-Mortem (Required for bug fix PRs)

N/A: scheduled catalog maintenance, not a bug fix.

Specialist Review Evidence (Required for agent-authored PRs)

• All local reviewers completed and findings addressed before merge
• If any reviewer did NOT complete: needs-human-review label added and merge deferred to human — N/A: every reviewer completed

Reviewer	Status	Outcome
task-completion-validator	PASS	Checks A–F passed; implementation and evidence match the task
constitution-validator	PASS	No constitutional violations; model IDs are intentional catalog data
test-engineer	PASS	Focused static-catalog regression coverage is sufficient

Exceptions (If any)

• Scope: None
• Rationale: N/A
• Expiration: N/A

Agent Preflight (Required)

• Preflight completed before code changes

Classification

• external-api-change
• cross-component-change
• business-logic-change
• public-surface-change
• docs-sync-change
• security-sensitive-change
• ui-change
• infra-change

External References

Official documentation was checked before coding: https://platform.claude.com/docs/en/about-claude/models/overview, https://developers.openai.com/api/docs/models, https://ai.google.dev/gemini-api/docs/models, https://docs.mistral.ai/models, and https://models.dev/api.json.

Codebase Impact Analysis

packages/shared/src/model-catalog.ts static OpenCode fallback data and packages/shared/tests/unit/model-catalog-alternative-providers.test.ts regression coverage.

Documentation & Specs

N/A: no user-facing contract or workflow changed; the research and validation record is archived under tasks/archive/.

Constitution & Risk Check

Checked Principle XI/no-hardcoded-values and the full project constitution. The hardcoded values here are the intentional static catalog payload. Primary risk is provider churn; source-based regression coverage and weekly refresh scope mitigate it.

[codspeed-hq]

Merging this PR will not alter performance

✅ 6 untouched benchmarks

---

_{Comparing sam/weekly-model-catalog-update-t59j22 (39dec12) with main (44adc7e)}

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud