Updates the canonical Anthropic Opus model in every fallback chain
(sisyphus, oracle, prometheus, metis, momus, visual-engineering,
ultrabrain, deep, artistry, unspecified-high), the unspecified-high
category default, the think-mode HIGH_VARIANT_MAP, the Claude Code
alias map, the claude-thinking legacy alias, the context-limit GA
regex, and event.ts fallback strings.
Widens supportsCachedAnthropicLimit to accept both claude-*-4-6 and
claude-*-4-7 so the 1M context cache still applies across the bump.
Regenerates the bundled model-capabilities snapshot from models.dev
and the model-fallback snapshot to match the new source output.
- Update AGENTS.md header with current date and commit
- Update runtime-fallback test to reflect quota STOP classification
- Release notes drafted in .sisyphus/drafts/release-notes-v3.16.0.md
Remove quota exhaustion patterns from RETRYABLE_ERROR_PATTERNS:
- 'usage limit reached' patterns (lines 30, 32)
- 'insufficient credits' pattern (line 37)
- 'credit balance too low' pattern (line 38)
These errors indicate permanent quota exhaustion, not temporary
rate limits. They are already handled by classifyErrorType() which returns
'quota_exceeded', and isRetryableError() properly stops on these
unless there's an explicit auto-retry signal.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-openagent)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
Display names with parentheses like 'Atlas (Plan Executor)' cause HTTP
header validation errors in x-opencode-agent-name. This was blocking
Atlas/Prometheus from working via /start-work and auto-retry.
Changes:
- Display names: parens -> dashes ('Atlas - Plan Executor')
- Hooks (start-work, no-hephaestus-non-gpt, no-sisyphus-gpt): use
config keys ('atlas', 'sisyphus', 'hephaestus') for agent API fields
- auto-retry: use config key instead of display name for promptAsync
- agent-override-protection: handle dash-suffix normalization
- Updated all test expectations to match new format
Closes#3138
Bug fixes:
1. OAuth token refresh (#3149): buildHttpRequestInit() now attempts silent refresh
via refresh_token before triggering full browser re-auth. Added refresh() method
to McpOAuthProvider. Includes test isolation fix for discovery mock.
2. Quota error STOP (#3126): Added STOP_MESSAGE_PATTERNS in model-error-classifier
that take precedence over RETRYABLE_MESSAGE_PATTERNS. Message-only quota errors
now non-retryable. Runtime-fallback: quota_exceeded with 'retrying in' signal
still triggers fallback (provider-managed auto-retry). Restored removed patterns.
3. Compaction loop (#3127): MAX_RECOVERY_ATTEMPTS=3 cap + additional suppression
guard from opencode session in degradation monitor.
Also: refactored extractAutoRetrySignal to auto-retry-signal.ts, new regression
tests for quota classifier and compaction degradation monitor.
- Add afterAll(() => { mock.restore() }) to 52 test files missing cleanup
- Rewrite create-tool-guard-hooks.test.ts to use spyOn instead of barrel mock
- Fix skill-mcp-manager OAuth tests with missing mockTokens/mockLogin definitions
- Fix start-work hook: show worktree active block on resume with existing worktree_path
- Extract createWorktreeActiveBlock to worktree-block.ts to avoid circular import
- Replace 80-line isolated test runner CI config with single `bun test` command
When a user cancels a generation (ESC x2), all three idle hooks could
enter a permanently broken state:
1. **todo-continuation-enforcer**: consecutiveFailures accumulated from
abort-caused promptAsync failures, eventually hitting MAX_CONSECUTIVE_FAILURES
and permanently stopping continuation injection.
2. **unstable-agent-babysitter**: no abort awareness at all — would keep
firing reminders after user cancelled the session.
3. **runtime-fallback**: retry dedupe keys and pending fallback state
persisted across cancellation, blocking legitimate error recovery.
Fix:
- Add shared `isAbortError()` utility for consistent abort detection
- Reset consecutiveFailures and clear stale state on AbortError in all hooks
- Track `lastCancelledAt` in todo-continuation-enforcer for abort window
- Add abort-awareness to unstable-agent-babysitter (skip if recently cancelled)
- Clear runtime-fallback retry state on abort errors
Tests: 61 pass, 0 fail across all 3 affected hook test suites.
Closes#2984
- background_output: snapshot read cursor before consuming, restore on
/undo message removal so re-reads return data (fixes#2915)
- MCP loader: preserve oauth field in transformMcpServer, add scope/
projectPath filtering so local-scoped MCPs only load in matching
directories (fixes#2917)
- runtime-fallback: add 'reached your usage limit' to retryable error
patterns so quota exhaustion triggers model fallback (fixes#2918)
Verified: bun test (4606 pass / 0 fail), tsc --noEmit clean
model_not_supported errors from providers (e.g. OpenAI returning
{"error": {"code": "model_not_supported"}}) were not recognized as
retryable. Subagents would silently fail with no response, hanging the
parent session.
Fix:
- Add "model_not_supported", "model not supported", "model is not
supported" to RETRYABLE_MESSAGE_PATTERNS in model-error-classifier.ts
- Add regex patterns to RETRYABLE_ERROR_PATTERNS in
runtime-fallback/constants.ts to match "model ... is ... not ...
supported" with flexible spacing
- Add regression test covering all three variations
Now model_not_supported errors trigger the normal fallback chain instead
of silent failure.
Add support for object-style entries in fallback_models arrays, enabling
per-model configuration of variant, reasoningEffort, temperature, top_p,
maxTokens, and thinking settings.
- Zod schema for FallbackModelObject with full validation
- normalizeFallbackModels() and flattenToFallbackModelStrings() utilities
- Provider-agnostic model resolution pipeline with fallback chain
- Session prompt params state management
- Fallback chain construction with prefix-match lookup
- Integration across delegate-task, background-agent, and plugin layers
The test name claimed it exercised RETRYABLE_ERROR_PATTERNS directly,
but classifyErrorType actually matches 'payment required' via the
quota_exceeded path first. Rename to 'detects payment required errors
as retryable' to accurately describe end-to-end behavior.
- error-classifier: pre-compile default retry pattern regex
- think-mode/detector: combine multilingual patterns into single regex
- parser: skip redundant toLowerCase on pre-lowered keywords
- edit-operations: use fast arraysEqual instead of JSON comparison
- hash-computation: optimize streaming line extraction with index tracking
When runtime fallback switches to a different model, the agent's
configured variant and reasoningEffort were lost because
buildRetryModelPayload only extracted variant from the fallback
model string itself.
Now buildRetryModelPayload accepts optional agentSettings and uses
the agent's variant as fallback when the model string doesn't
include one. reasoningEffort is also passed through.
🤖 Generated with assistance of [OhMyOpenCode](https://github.com/code-yeongyu/oh-my-opencode)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The nullish-coalescing chain could stop at a non-numeric value (e.g.
status: "error"), preventing deeper nested numeric statusCode values
from being reached. Switch to Array.find() with a type guard to always
select the first numeric value.
Adds 11 tests for extractStatusCode covering: top-level, nested
(data/error/cause), non-numeric skip, fallback to regex, and
precedence.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
AI SDK wraps HTTP status codes inside error.error.statusCode (e.g., AI_APICallError). The current extractStatusCode only checks the top level, missing these nested codes.
This caused runtime-fallback to skip retryable errors like 400, 500, 504 because it couldn't find the status code.
Fixes#2617
When Gemini returns a quota exhausted error, OpenCode auto-retries and
fires session.status with type='retry'. The extractAutoRetrySignal
function requires BOTH 'retrying in' text AND a quota pattern to match,
but some providers (like Gemini) include only the error text in the
retry message without the 'retrying in' phrase.
Since status.type='retry' already confirms this is a retry event, the
fix adds a fallback check: if extractAutoRetrySignal fails, check the
message directly against RETRYABLE_ERROR_PATTERNS. This ensures quota
errors like 'exhausted your capacity' trigger the fallback chain even
when the retry message format differs from expected.
Fixes#2454
setInterval for model availability monitoring was never cleared,
keeping the hook alive indefinitely with no dispose mechanism.
- Add dispose() method to RuntimeFallbackHook that clears interval
- Track intervalId in hook state for cleanup
- Export dispose in hook return type
Tests: 3 pass, 10 expects
YeonGyu-Kim