Mock constants directly in cache.test to avoid transitive module-cache reuse when importing cache.ts. This removes Date.now query cache-busting and makes the test deterministic across runs.
Align version lookup, invalidation, and bun install with OpenCode's cache directory so updates target the loaded plugin location. Keep dependency declarations intact during invalidation so auto-update can reinstall instead of converging to uninstall.
Sisyphus can now fall back through Kimi and OpenAI models when Claude
is unavailable, enabling OpenAI-only users to use Sisyphus directly
instead of being redirected to Hephaestus.
Runtime chain: claude-opus-4-6 max → k2p5 → kimi-k2.5 → gpt-5.4 medium → glm-5 → big-pickle
CLI chain: claude-opus-4-6 max → k2p5 → gpt-5.4 medium → glm-5
Align runtime defaults, tests, docs, and generated artifacts with the newer GPT-5.4 baseline. Keep think-mode and prompt-routing expectations consistent after the model version bump.
Add MANUAL_QA_MANDATE sections to all three ultrawork prompts (default,
GPT, Gemini). Agents must now define acceptance criteria in TODO/Task items
before implementation, then execute manual QA themselves after completing
work. lsp_diagnostics alone is explicitly called out as insufficient since
it only catches type errors, not functional bugs.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
Replace real setTimeout(7000) with fake timer interception in atlas
retry tests (35s -> 227ms). Add missing opencode provider to glm-5
fallback in unspecified-high category.
When atlas injects a boulder continuation via promptAsync() and the
model's response is immediately aborted (MessageAbortedError), OpenCode
fires a burst of session.idle events within milliseconds. Atlas blocks
all of them due to the 5-second cooldown. After the burst, OpenCode
stops generating session.idle events (it's state-change based, not
periodic), leaving the session stuck forever.
Fix: When cooldown blocks an idle event for a boulder session with an
incomplete plan, schedule a one-shot setTimeout (cooldown + 1s) to
re-attempt injection. The timer callback re-checks boulder state, plan
progress, and continuation-stopped flag before injecting. Only one timer
per session is allowed (deduped via pendingRetryTimer field). Timers are
cleaned up on session.deleted and session.compacted events.
When no worktree is specified in boulder, stop injecting 'Worktree Setup Required'
instructions. When worktree IS present, inject emphatic instructions ensuring the
agent and all subagents operate exclusively within the worktree directory.
Fix#2289
invalidatePackage() only removed the plugin from USER_CONFIG_DIR/node_modules/,
but bun may install it in CACHE_DIR/node_modules/ on some systems. This left a
stale copy behind, causing the startup toast to keep showing the old version even
after the auto-update completed successfully.
Now both candidate locations are checked and removed so the reinstalled version
is loaded on the next restart.
compactedSessions permanently blocked re-compaction after first success,
causing unbounded context growth (e.g. 500k on Kimi K2.5 with 256k limit).
- Clear compactedSessions flag on new message.updated so compaction can
re-trigger when context exceeds threshold again
- Use modelContextLimitsCache for model-specific context limits instead
of always falling back to 200k for non-Anthropic providers
Previously decoded entire image buffer to read headers. Now slices base64
to 32KB prefix before decoding — sufficient for PNG/GIF/WebP/JPEG headers.
Dramatically reduces memory allocation for large images.
Intercepts Read tool output with image attachments and resizes to comply with Anthropic API limits (≤1568px long edge, ≤5MB). Only activates for Anthropic provider sessions and appends resize metadata (original/new resolution, token count) to tool output.
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
acamq