[devin-ai-integration]

Summary

Part of a coverage push across COG-GTM repos. Measured baseline with pytest -m 'not pricy' --cov=agentlab (MiniWoB server running): 42% statement coverage. This PR adds 33 focused unit tests for pure-logic modules that previously had large untested areas, raising overall coverage to 44% (100 passed, 15 skipped; the pre-existing test_llm_error_no_success failure needs a real OpenAI key and is unaffected).

• tests/llm/test_llm_utils_extra.py — messages_to_dict, extract_html_tags, parse_html_tags(_raise) (incl. optional keys, merge_multiple, retry messages), retry_multiple (success/retry/exhaustion via a mocked chat), generic_call_api_with_retries (valid + invalid-then-valid responses), image_to_{jpg,png}_base64_url, BaseMessage/SystemMessage/HumanMessage/AIMessage, Discussion (append/to_openai/merge/add_content).
• tests/llm/test_prompt_templates.py — PromptTemplate.format_message/construct_prompt incl. error paths, get_prompt_template lookup.
• tests/agents/test_agent_utils.py — draw_mouse_pointer, draw_click_indicator, draw_arrowhead (pixel assertions), zoom_webpage (mocked Page, invalid-factor ValueError).

Formatted with black. Reaching 70% overall requires covering the large untested modules (agent_xray.py 713 stmts Gradio UI, tool_use_agent.py 396, inspect_results.py 309, study.py 201…) — proposed as follow-up work.

Link to Devin session: https://app.devin.ai/sessions/176aa77533c34040b8d9dd36595343fa
Requested by: @SachetCognition

---

Devin Review

Status	Commit
⚪ Not started	—

Run Devin Review

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment, CI, and merge conflict monitoring