mirror of
https://github.com/tiennm99/ccs.git
synced 2026-10-11 03:13:12 +00:00
Kenneth Wong
3583b544d0
Register `claude-fable-5-1` in the CLIProxy model catalog, the dashboard
catalog, and the usage pricing registry.
Pricing is taken from Anthropic's official pricing page:
- Fable 5.1 base rates are $10/$50 per MTok with a $12.50 5m cache write,
matching Fable 5.
- Cache hits bill at 0.025x base input ($0.25/MTok) rather than the
standard 0.1x multiplier. Anthropic applies that reduced rate only to
Fable 5.1 and Mythos 5.1, so this entry cannot derive its cache rates
from CACHE_READ_MULTIPLIER.
This also corrects Claude Sonnet 5 from $3/$15 to $2/$10 (cache write
$2.50, cache read $0.20). The launch introductory rate became the
standard price and the increase scheduled for 2026-09-01 was cancelled,
so the previous entry over-reported Sonnet 5 usage cost by 50%.
Thinking on Fable 5.1 is always on and can only be steered through
effort levels, so the catalog entry exposes the same `low`..`max` level
surface as Fable 5 and Opus 5 instead of a manual token budget.
The GitHub Copilot catalog is deliberately left unchanged, since Copilot
availability for Fable 5.1 is not verified; surfacing it there would
offer a model the backend may reject.
Regenerates docs/reports/hardening-inventory.{json,md} so the ci-parity
gate matches the source tree.
CCS Test Suite
Root TypeScript and JavaScript tests run with Bun's test runner. Native shell, PowerShell, Docker, and standalone probes cover contracts that need a real platform or process boundary.
Ownership
| Area | Path | Use for |
|---|---|---|
| Unit | unit/ |
Focused module and command behavior |
| Integration | integration/ |
Cross-module, process, proxy, auth, and web-server behavior |
| End to end | e2e/ |
Packaged CLI workflows |
| npm package | npm/ |
Installation, exports, and package behavior |
| Native | native/ |
Unix and Windows shell behavior |
| Docker | docker/ |
Compose and stable network/service contracts |
| Documentation | docs/ |
Repository documentation invariants |
| Shared support | shared/ |
Fixtures and helpers reused by suites |
| Mocks | mocks/ |
Bounded test doubles and fixtures |
Some source domains also keep focused tests in src/**/__tests__/. Follow the
nearest established pattern and avoid moving tests solely for taxonomy.
Commands
Commands are defined in ../package.json. Bucket membership
and execution live in
../scripts/run-test-bucket.js.
bun run test:fast # Fast Bun test bucket
bun run test:slow # Slow Bun test bucket
bun run test:all # All root non-e2e Bun test buckets
bun run test:unit # tests/unit
bun run test:npm # tests/npm
bun run test:native # Native Unix edge-case script
bun run test:e2e # tests/e2e with fail-fast and extended timeout
bun run test # Build, then test:all
For the normal contributor gate:
bun run format
bun run lint:fix
bun run validate
For the closest local equivalent to PR CI:
bun run validate:ci-parity
Dashboard tests use Vitest and are documented in
../ui/README.md.
Test Isolation
- Set
CCS_HOMEto a temporary directory. - Never read or modify a contributor's real
~/.ccs/or~/.claude/. - Use
getCcsDir()from../src/utils/config-manager.tsfor CCS paths. - Keep fixtures deterministic and free of credentials or private account data.
- Use real behavior at the boundary under test; do not weaken assertions to hide regressions.
Adding Coverage
- Add a focused unit test for isolated logic.
- Add integration coverage when behavior crosses modules, processes, HTTP, or persistence boundaries.
- Add e2e coverage when command routing or packaged CLI behavior is the contract.
- Add native or Docker coverage only when platform/runtime behavior cannot be represented faithfully in Bun tests.
- Run the smallest relevant command first, then broaden to the contributor or CI-parity gate when shared contracts changed.