feat(extraction): add mode-specific prompt hierarchy and guardrails (#678)

[aredenba-rh]

Summary

• introduce structured resolved skill packs for extraction modes with explicit system prompt, prompt hierarchy, guardrails, and default mode-specific skills
• add a tracer-bullet extraction skill override repository and wire dependency injection so session creation resolves mode-aware configuration
• persist resolved agent configuration into new session runtime context for stable downstream agent runtime usage

Testing

• cd src/api && uv run pytest tests/unit/extraction/application/test_skill_resolution_service.py tests/unit/extraction/application/test_agent_session_service.py tests/unit/extraction/presentation/test_routes.py tests/unit/extraction/test_architecture.py -q

Risks

• low: extraction session runtime_context shape now includes agent_configuration; existing sessions without this key remain supported

Closes #678

Made with Cursor

[coderabbitai]

Important

Review skipped

Auto reviews are disabled on base/target branches other than the default branch.

Please check the settings in the CodeRabbit UI or the .coderabbit.yaml file in this repository. To trigger a single review, invoke the @coderabbitai review command.

⚙️ Run configuration

Configuration used: Central YAML (base), Organization UI (inherited)

Review profile: ASSERTIVE

Plan: Enterprise

Run ID: 9e1b7efd-832e-49da-ac28-7bf513f2f42f

You can disable this status message by setting the reviews.review_status to false in the CodeRabbit configuration file.

Use the checkbox below for a quick retry:

• 🔍 Trigger review

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Commit unit tests in branch issue-678-mode-specific-prompts-and-skill-packs

---

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