fix(slopguard): update default review models and effort - #123
Conversation
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
![slopzapper[bot]](https://avatars.githubusercontent.com/in/4527847?s=60&v=4){kind=small}
There was a problem hiding this comment.
🟡 Changes recommended
The updated configuration documentation is misleading for Cursor because the adapter rejects any explicitly configured reasoning_effort, so the docs should clarify this exception.
Once you've addressed the issues Copilot identified, you can request another Copilot review.
Pull request overview
Updates Slopguard’s built-in default provider models and default reasoning-effort behavior to match the requested Astra/Fable/Grok defaults, and aligns documentation + contract tests accordingly.
Changes:
- Switch default models: Codex →
gpt-6-astra, Claude →claude-fable-5-1, Cursor →cursor-grok-4.6-high(drop “fast”). - Change default reasoning effort to
mediumfor Claude (Codex remainsmedium; Cursor/Grok remainhigh). - Update provider docs and adjust tests (including live smoke tests) to reflect the new defaults.
File summaries
| File | Description |
|---|---|
| skills/slopguard/references/providers.md | Updates skill provider-default table + fallback guidance. |
| skills/slopguard/references/configuration.md | Updates documented default model + reasoning-effort defaults. |
| cli/slopguard/internal/provider/types.go | Changes compiled-in default model constants. |
| cli/slopguard/internal/provider/cursor_test.go | Updates Cursor default-model expectation. |
| cli/slopguard/internal/provider/codex_live_test.go | Forces live Codex smoke to run at medium effort with default model. |
| cli/slopguard/internal/provider/claude_live_test.go | Forces live Claude smoke to run at medium effort with default model. |
| cli/slopguard/internal/config/load.go | Applies medium default effort for Claude (alongside Codex). |
| cli/slopguard/internal/config/config_test.go | Updates config default-effort tests for Claude. |
| cli/slopguard/docs/engines/README.md | Updates engine defaults table. |
| cli/slopguard/docs/engines/cursor.md | Updates Cursor default model reference. |
| cli/slopguard/docs/engines/codex.md | Updates Codex default model reference + fallback note. |
| cli/slopguard/docs/engines/claude-code.md | Updates Claude default model/effort + fallback note. |
| cli/slopguard/cmd/slopguard/main_test.go | Updates CLI config command default-effort expectations for Claude. |
Review details
- Files reviewed: 13/13 changed files
- Comments generated: 1
- Review effort level: Lite
💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.
|
Problem
Slopguard used Sol for Codex, Opus at high effort for Claude, and the fast Grok variant for Cursor. The requested defaults are Astra medium, Fable 5.1 medium, and Grok 4.6 high without fast mode.
Closes #122.
Solution
Set Codex to
gpt-6-astra, Claude toclaude-fable-5-1at medium effort, and Cursor tocursor-grok-4.6-high. Grok remainsgrok-4.6at high effort. Update the provider documentation and existing contract tests.Availability was checked during this change. Runtime model selection remains fixed; accounts without access can explicitly select
gpt-5.6-solorclaude-opus-5, both at medium effort.Proof
Authenticated Slopguard adapter smoke tests passed with Astra medium and Fable 5.1 medium through the installed CLIs. The installed Cursor model catalog lists
cursor-grok-4.6-high.