* fix(ralph-loop): skip user messages in transcript completion detection (#622)
The transcript-based completion detection was searching the entire JSONL
file for <promise>DONE</promise>, including user message entries. The
RALPH_LOOP_TEMPLATE instructional text contains this literal pattern,
which gets recorded as a user message, causing false positive completion
detection on every iteration. This made the loop always terminate at
iteration 1.
Fix: Parse JSONL entries line-by-line and skip entries with type 'user'
so only tool_result/assistant entries are checked for the completion
promise. Also remove the hardcoded <promise>DONE</promise> from the
template exit conditions as defense-in-depth.
* chore: changes by sisyphus-dev-ai
---------
Co-authored-by: sisyphus-dev-ai <sisyphus-dev-ai@users.noreply.github.com>
MODEL_TO_CATEGORY_MAP was missing the mapping for google/gemini-3-flash
to the 'writing' category. Users who had configured agents with
model: 'google/gemini-3-flash' would not get auto-migrated to
category: 'writing'.
Ref: PR #1057 review comment
Co-authored-by: justsisyphus <justsisyphus@users.noreply.github.com>
On Windows with Bun v1.3.5 and earlier, spawning LSP servers causes
a segmentation fault crash. This is a known Bun bug fixed in v1.3.6.
Added version check before LSP server spawn that:
- Detects Windows + affected Bun versions (< 1.3.6)
- Throws helpful error with upgrade instructions instead of crashing
- References the Bun issue for users to track
Closes#1047
Add thinking mode support for Z.AI's GLM-4.7 model via the zai-coding-plan provider.
Changes:
- Add zai-coding-plan to THINKING_CONFIGS with extra_body.thinking config
- Add glm pattern to THINKING_CAPABLE_MODELS
- Add comprehensive tests for GLM thinking mode
GLM-4.7 uses OpenAI-compatible API with extra_body wrapper for thinking:
- thinking.type: 'enabled' or 'disabled'
- thinking.clear_thinking: false (Preserved Thinking mode)
Closes#1030
- Remove invalid Pick<Plugin> type usage
- Add explicit input/output type annotations
- Add comprehensive test suite (5 tests)
- Tests verify truncation at 30 chars with '...' suffix
When AI generates AskUserQuestion tool calls with option labels longer
than 30 characters, opencode validation rejects them with "too_big" error.
This fix adds a pre-tool-use hook that automatically truncates labels
to 30 characters (with "..." suffix) before the validation occurs.
Fixes the error:
"The question tool was called with invalid arguments: expected string
to have <=30 characters"
Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
When a later-loaded MCP config (e.g., .claude/.mcp.json) marks a server
as disabled, it now properly removes that server from both the servers
object and loadedServers array.
Previously, disabled servers were only skipped during loading, which
meant they wouldn't override servers loaded from earlier configs. This
made it impossible to disable project-level MCPs using local overrides.
Now the disabled flag works as expected: local configs can disable
servers defined in project or user configs.
This ensures that commands defined in src/features/builtin-commands/commands.ts
(like /start-work, /refactor, /init-deep) are visible to the slashcommand tool
and the agent. Previously, only markdown-based commands were discovered.
The background_task config (providerConcurrency, modelConcurrency, etc.)
was not being passed to BackgroundManager, causing all models to use
the hardcoded default limit of 5 instead of user-configured values.
- Add 'compaction' to DEFAULT_SKIP_AGENTS
- Skip compaction agent messages when resolving agent info
- Skip injection when compaction occurred but no real agent resolved
- Replace cooldown-based approach with agent-based filtering
OpenCode downloads LSP servers (like clangd) to ~/.local/share/opencode/bin,
but isServerInstalled() only checked ~/.config/opencode/bin. This caused
LSP tools to report servers as 'not installed' even when OpenCode had
successfully downloaded them.
Add ~/.local/share/opencode/bin to the detection paths to match OpenCode's
actual behavior.
Co-authored-by: yimingll <yimingll@users.noreply.github.com>
* fix(skill): enforce agent restriction in createSkillTool
Ultraworked with [Sisyphus](https://github.com/code-yeongyu/oh-my-opencode)
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
* fix(skill): block restricted skills when agent context missing
Addresses cubic review feedback: previously agent-restricted skills
could be invoked when ctx or ctx.agent was undefined because the
guard only ran when ctx?.agent was truthy.
Changed condition from:
skill.definition.agent && ctx?.agent && skill.definition.agent !== ctx.agent
To:
skill.definition.agent && (!ctx?.agent || skill.definition.agent !== ctx.agent)
This ensures restricted skills are blocked unless the exact matching
agent is present in the context.
---------
Co-authored-by: justsisyphus <justsisyphus@users.noreply.github.com>
Co-authored-by: Sisyphus <clio-agent@sisyphuslabs.ai>
When git commands fail (e.g., not in a repo, invalid HEAD), git outputs
help text to stderr. Without explicit stdio option, execSync inherits
the parent's stdio causing help text to appear on terminal during
delegate_task execution.
Add stdio: ['pipe', 'pipe', 'pipe'] to capture stderr instead of
letting it leak to terminal.
When availableModels is empty (no cache in CI), use the first entry
from fallbackChain directly instead of falling back to systemDefault.
This ensures categories and agents use their configured models even
when the model cache file doesn't exist.
Fixes:
- model-resolution check returning 'warn' instead of 'pass' in CI
- DEFAULT_CATEGORIES not being used when no cache available
- Unstable agent detection failing (models falling back to non-gemini)
resolveModelWithFallback() returns entry-specific variant but it was being
ignored. Agents like oracle now correctly get variant 'high' from their
fallback chain entry instead of undefined.
justsisyphus