[mitchins]

Summary

• add provider-wide and exact-model system prompt transforms for text chat completions
• support append, prepend, and replace modes with canonical model resolution
• preserve router-owned /no_think, skip structured system content safely, and document configuration

Verification

• pytest -q — 1070 passed, 9 skipped
• ruff check smolrouter/ tests/unit/test_system_prompts.py
• vulture smolrouter/ --min-confidence=80 --exclude .venv --exclude venv --exclude .conda --exclude .cache
• git diff --check

Peer-reviewed before and after implementation by GPT-5.6-Sol.

Summary by CodeRabbit

• New Features

  • Added configurable system-prompt transformations for OpenAI-compatible chat requests.
  • Supports appending, prepending, or replacing system prompts.
  • Supports provider-wide defaults and exact per-model overrides.
  • Preserves supported router markers and message ordering.

• Behavior Updates

  • Transformations are skipped for audio-output requests, Responses requests, and unsupported structured messages.
  • Requests remain unchanged when transformations cannot be safely applied.

• Documentation

  • Added configuration guidance covering supported modes, precedence, opt-out behavior, and request limitations.

[chatgpt-codex-connector]

Codex Review Summary

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	✅ Completed 2026-09-12T22:22:37.544505Z	09def22	PR opened

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: defaults

Review profile: CHILL

Plan: Essentials

Run ID: e7889326-b008-402d-b2a5-92a19e5a9715

📥 Commits

Reviewing files that changed from the base of the PR and between 09def22 and e5d678f.

📒 Files selected for processing (1)

• pyproject.toml

Included review availability: 1 review is currently available. Your included PR review attempts over the past 7 days set your current allowance at 3 reviews per hour.

---

📝 Walkthrough

Walkthrough

This change adds validated provider-wide and model-specific system prompt policies. Eligible OpenAI chat-completions requests receive immutable prompt transformations. Audio, Responses, and unsupported message inputs remain unchanged.

Changes

System prompt policy

Layer / File(s)	Summary
Policy configuration and resolution smolrouter/interfaces.py, tests/unit/test_system_prompts.py	Adds immutable configuration validation, provider-wide and exact-model coercion, override precedence, opt-out handling, and YAML coverage.
System message transformation smolrouter/system_prompts.py, tests/unit/test_system_prompts.py	Adds append, prepend, and replace modes. The logic copies messages, preserves /no_think markers, rejects unsupported shapes, and reports skip reasons.
Chat-completions policy application smolrouter/mediator.py, smolrouter/providers.py, tests/unit/test_system_prompts.py, README.md, pyproject.toml	Applies policies before provider dispatch, preserves caller-owned payloads, skips audio and Responses requests, updates provider error logging, documents configuration, and enables Ruff checks.

Priority: ➖ Normal

Estimated code review effort: 3 (Moderate) | ~25 minutes

Change: Feature

Sequence Diagram(s)

sequenceDiagram
  participant Client
  participant route_request
  participant ProviderConfig
  participant transform_system_prompt_messages
  participant SelectedProvider
  Client->>route_request: submit chat-completions payload
  route_request->>ProviderConfig: resolve policy for model
  ProviderConfig-->>route_request: return policy or no policy
  route_request->>transform_system_prompt_messages: transform copied messages
  transform_system_prompt_messages-->>route_request: return transformed messages
  route_request->>SelectedProvider: dispatch request

Loading

Merge Risk: ⚪ Minimal · up to e5d67

This configuration-only change introduces no identified merge-blocking risk.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 30.00% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 30 functions across 5 files. (1 skipped: …	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title clearly and concisely describes the main change: adding model-aware system prompt transforms.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.

Full details: Docstring Coverage

Explanation

Docstring coverage is 30.00% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 30 functions across 5 files. (1 skipped: 1 unsupported.)

• Fix all pre-merge checks with AI

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch codex/model-aware-system-prompt-transforms

---

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

[codecov]

Codecov Report

❌ Patch coverage is 87.14286% with 18 lines in your changes missing coverage. Please review.
✅ Project coverage is 76.94%. Comparing base (7568c8f) to head (e5d678f).

Files with missing lines	Patch %	Lines
smolrouter/mediator.py	75.00%	4 Missing and 4 partials ⚠️
smolrouter/interfaces.py	86.66%	3 Missing and 3 partials ⚠️
smolrouter/system_prompts.py	93.33%	2 Missing and 2 partials ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #40      +/-   ##
==========================================
+ Coverage   76.82%   76.94%   +0.12%     
==========================================
  Files          38       39       +1     
  Lines       10902    11041     +139     
  Branches     1685     1720      +35     
==========================================
+ Hits         8375     8496     +121     
- Misses       1930     1939       +9     
- Partials      597      606       +9     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
93.7% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[mitchins]

@coderabbitai review quota refreshed, review

[coderabbitai]

@mitchins I will review the changes in #40.

✅ Action performed

Review finished.

Note: CodeRabbit is an incremental review system and does not re-review already reviewed commits. This command is applicable only when automatic reviews are paused.