Conversation
…cepts it resolve_confirmed_agent requires autonomous_args() to return Some before dispatching, as a safety gate against agents with no known permission-bypass flag. opencode's entry was None, so items assigned to it were silently skipped by the discovery tick forever (dispatched 0, skipped 1 every tick). opencode run --help confirms --auto: "auto-approve permissions that are not explicitly denied" -- the same role --dangerously-skip-permissions plays for claude-code. Mapped it in autonomous_args, flipped the now-wrong resolve_confirmed_agent_rejects_opencode test to resolve_confirmed_agent_accepts_opencode, and switched unconfirmed_agent_gets_skipped_not_dispatched's example agent from opencode (now confirmed) to cursor (still unmapped). Agentflare-Agent: claude-code Agentflare-Branch: task/103 Agentflare-Item: 103
|
Warning Review limit reachedYou’ve reached a temporary PR review limit under our Fair Usage Limits Policy. Next review available in: 28 minutes Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available. How can I continue?After more reviews become available, a review can be triggered using the To avoid repeated limits, reduce automatic review volume by pausing incremental auto-reviews earlier, using label-based review opt-in, excluding WIP or generated PR titles, or requesting reviews manually when the PR is ready. If your team needs uninterrupted high-volume reviews, an organization admin can enable usage-based reviews. How do review limits work?CodeRabbit enforces per-developer PR review limits for each organization. Most developers receive the normal plan review availability. For paid Pro and Pro+ PR reviews, CodeRabbit uses adaptive limits for sustained high-volume activity. When a developer's recent PR review activity reaches the 95th percentile or higher among CodeRabbit users, additional reviews become available more gradually as earlier reviews age out of the rolling window. Please refer docs for additional details. Review details⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (4)
📝 WalkthroughWalkthroughThe agent registry now returns ChangesOpencode autonomy
Estimated code review effort: 2 (Simple) | ~10 minutes 🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
✨ Finishing Touches📝 Generate docstrings
🧪 Generate unit tests (beta)
Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
🧹 Nitpick comments (1)
src/supervisor.rs (1)
596-600: 🎯 Functional Correctness | 🔵 Trivial | ⚡ Quick winCover the exact autonomous argument.
This test verifies agent recognition only. It would also pass if
Agent::Opencodereturned an incorrect non-empty argument. Add an assertion forautonomous_args(Agent::Opencode) == Some(&["--auto"]), or verify that the dispatched command containsrunand--auto.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In `@src/supervisor.rs` around lines 596 - 600, Extend resolve_confirmed_agent_accepts_opencode to assert the exact autonomous arguments for Agent::Opencode, verifying autonomous_args returns Some(&["--auto"]). Keep the existing agent-recognition assertion and use the established autonomous_args helper.
🤖 Prompt for all review comments with AI agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.
Nitpick comments:
In `@src/supervisor.rs`:
- Around line 596-600: Extend resolve_confirmed_agent_accepts_opencode to assert
the exact autonomous arguments for Agent::Opencode, verifying autonomous_args
returns Some(&["--auto"]). Keep the existing agent-recognition assertion and use
the established autonomous_args helper.
ℹ️ Review info
⚙️ Run configuration
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
Run ID: 7db20c01-c893-4f72-97ed-c7567d93ed7d
📒 Files selected for processing (2)
crates/agent-registry/src/registry.rssrc/supervisor.rs
…was unconfirmed Same class of breakage as the supervisor.rs test fixed in the prior commit -- src/quota/decide.rs's decide() calls the same supervisor::resolve_confirmed_agent, and this test used opencode as its 'definitely unconfirmed' example. Switched to cursor, matching the supervisor.rs fix. Agentflare-Agent: claude-code Agentflare-Branch: task/103 Agentflare-Item: 103
…gh to --model
Adds WorkArgs.model (agentflare work --model <name>, CLI) and a matching
metadata.model item field (read by supervisor::item_model_override) for
autonomous dispatch -- set via handoff/item update's metadata, e.g.
{"model": "anthropic/claude-sonnet-5"}. Threaded through enqueue_work_job's
job args and WorkItemExecutor::execute the same way folder_path already is,
then appended in build_extra_args as --model <name> for any confirmed agent
(claude-code and opencode's --help both confirm the same flag spelling).
No allowlist: passed straight through with no validation against a hardcoded
model list, since catalogs change too often to keep in sync and the
underlying agent CLI already rejects an unknown name itself.
Also splits supervisor.rs's test module out to supervisor_tests.rs (pure
code motion, #[path] attribute) -- the new tests pushed it over the repo's
1500-line LOC gate.
Agentflare-Agent: claude-code
Agentflare-Branch: task/103
Agentflare-Item: 103
1 participant
Summary by CodeRabbit
New Features
Bug Fixes