feat(plugins): apply augment_prompt on Direct-API/Codex backends

[OBenner]

Problem

Plugin prompt augmentation (augment_prompt) reached the Claude SDK backend only. create_client() appends enabled plugins' contributions to the system prompt — but the Direct-API/Codex path builds its session with an empty system prompt and carries the agent's instructions in the message. So contributions from rules-steering-compiler, codebase-intelligence, skill-pack-runtime, etc. were silently absent from every non-claude build.

Change

Append the same augmentations to the agent message on the Direct-API path, right after the provider is resolved (coder.py). Extracted as a small helper _augment_direct_api_prompt for testability:

• Guarded on provider name — no-op for claude (it already augments via create_client), so no double-application.
• Failures degrade to the un-augmented prompt (never breaks a build).
• Covers the primary autonomous planner + coder direct-API flow.

Scope / follow-ups

• The follow-up planner (planner.run_followup_planner) uses a separate session builder — left as a focused follow-up.
• Docs interaction: companion PR feat(plugins): run pre_tool/post_tool hooks on the Direct-API runtime #417 (Direct-API tool hooks) adds a plugin-development note stating augment_prompt is "Claude SDK only." Once both land, that note's augment_prompt row flips to ✅ for Direct-API. Kept out of this PR to avoid a same-file merge conflict; will reconcile on merge.

Tests

4 new unit tests (test_coder_prompt_augmentation.py): claude is not augmented (spy asserts apply not called), openai/codex are augmented with the right args, and augmentation failure degrades to the base prompt. Ruff check + format clean.

🤖 Generated with Claude Code

[coderabbitai]

Warning

Review limit reached

@OBenner, you've reached your PR review limit, so we couldn't start this review.

Next review available in: 41 minutes

Enable usage-based reviews in Billing to review now. Otherwise, wait until the next included review is available.
You're only billed for reviews past your plan's rate limits ($0.25/file).

How can I continue?

After more reviews become available, a review can be triggered using the @coderabbitai review command as a PR comment. Alternatively, push new commits to this PR.

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 configuration

Configuration used: Path: .coderabbit.yaml

Review profile: ASSERTIVE

Plan: Pro

Run ID: 17425d3f-297f-41c5-a601-4483da918dce

📥 Commits

Reviewing files that changed from the base of the PR and between 4ea224f and 9bf1000.

📒 Files selected for processing (2)

• apps/backend/agents/coder.py
• tests/test_coder_prompt_augmentation.py

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch claude/augment-prompt-direct-api

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud