Skip to content
Permalink

Comparing changes

Choose two branches to see what’s changed or to start a new pull request. If you need to, you can also or learn more about diff comparisons.

Open a pull request

Create a new pull request by comparing changes across two branches. If you need to, you can also . Learn more about diff comparisons here.
base repository: onllm-dev/onWatch
Failed to load repositories. Confirm that selected base ref is valid, then try again.
Loading
base: v2.11.32
Choose a base ref
...
head repository: onllm-dev/onWatch
Failed to load repositories. Confirm that selected head ref is valid, then try again.
Loading
compare: v2.11.33
Choose a head ref
  • 1 commit
  • 38 files changed
  • 1 contributor

Commits on Apr 1, 2026

  1. feat: zero-429 Anthropic monitoring via statusline bridge, Codex Team… 

    … support, test safety

Anthropic Rate Limit Resilience:
- Statusline bridge: auto-configures Claude Code to write rate limit data
  to a shared file, eliminating API calls and 429 errors entirely
- Hybrid polling: statusline primary, API every N cycles for supplementary
  quotas (Sonnet, extra_usage) not available in statusline
- Claude Code detection guards BOTH OAuth refresh paths (proactive + 429
  bypass) to prevent burning Claude Code's one-time-use refresh token
- Backoff decay: failCount decrements on expiry and success, preventing
  infinite escalation
- Retry-After header support from OAuth 429 responses
- IsExpiringSoon zero-guard prevents spurious refreshes when expiry unknown
- Configurable via ANTHROPIC_SOURCE env var (auto/statusline/api) and
  Settings UI with DB persistence

Codex Team Support (#49):
- Composite external IDs (account_id:user_id) for per-user dedup
- Proactive refresh failure capping (pause after 3 consecutive failures)
- Credential layering: named profiles read from system auth.json when stale
- ParseIDTokenUserID extracts chatgpt_user_id from JWT claims

Dashboard & Menubar:
- Per-quota freshness labels (Live/API + age) on Anthropic cards
- Supplementary card in menubar separates API-sourced quotas visually
- Graph carry-forward prevents Sonnet from charting as 0% on statusline snapshots
- Source indicator in history API, detail modals, and session tables
- Provider-specific settings section in Settings > Providers tab

Test Safety:
- testMode flag prevents all keychain/keyring reads and writes during tests
- TestMain in api and agent packages enables testMode automatically
- Tests can never log out Claude Code again
- All OAuth tests use mock servers via SetOAuthURLForTest
- All tests override isClaudeCodeRunning to prevent real process detection

Closes #49
Refs: anthropics/claude-code#31637
    @prakersh
    prakersh committed Apr 1, 2026
    Configuration menu
    Copy the full SHA
    36af579 View commit details
    Browse the repository at this point in the history
Loading