feat: add agent markdown for better integration call

[mcllerena]

No description provided.

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 95.55%. Comparing base (2fc5da5) to head (91f891d).

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #142   +/-   ##
=======================================
  Coverage   95.55%   95.55%           
=======================================
  Files           8        8           
  Lines        1823     1823           
=======================================
  Hits         1742     1742           
  Misses         81       81           

☔ 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.

[pesap]

Requesting changes because the added file appears to use the wrong discovery filename for the intended agent integration. Please rename AGENT.md to AGENTS.md and keep the content scoped/compact per the AGENTS.md guidance source.

[pesap]

This should be AGENTS.md, not AGENT.md, for the standard agent-discovery convention. The provided source says the target is AGENTS.md files and its repo-level template is # AGENTS.md; official Codex docs likewise discover project instructions from AGENTS.md. Please rename this file to AGENTS.md so the intended integrations actually load it. Source: https://gist.github.com/IgorWarzocha/42ed1836aace7c44eb127f6194905820

[pesap]

Follow-up to my review: I should have compared the proposed AGENT.md content against the provided AGENTS.md guide, not only checked the discovery filename.

Reference: https://gist.github.com/IgorWarzocha/42ed1836aace7c44eb127f6194905820

Against that guide, I think this should be revised before merge:

• Rename AGENT.md to AGENTS.md; the referenced guide is specifically for discoverable AGENTS.md files.
• Compress the file substantially. The guide says normal repo/brownfield root AGENTS.md files should aim for ~20 lines, and that files over ~100 lines are probably docs/spec/runbook content. This file is currently 110 lines.
• Remove or compress duplicated/inferable material. The guide says not to restate what agents can infer from code, package/config files, folder names, or quick search. Several current sections repeat visible tooling or generic Python/project workflow guidance.
• Keep only repo-specific, behavior-changing guidance: concurrent-work safety, SQLite/model-processing performance sensitivity, verification exceptions, architecture seams, and non-obvious local conventions.

Suggested direction:

# AGENTS.md

- Assume concurrent human/agent work; never revert or overwrite changes you did not author.
- This repo processes model data into SQLite; watch query/transform cost, memory churn, and fixture representativeness.
- Prefer minimal, behavior-proven changes; add regression coverage for bug fixes and use existing fixtures under `tests/fixtures` or `tests/data` when practical.
- No breadcrumbs after deleting or moving code; remove the old code/comment instead of leaving relocation notes.
- Use `uv sync --all-groups` for setup. For changed areas, run the narrow relevant check first; before broad handoff use `uv run prek run --show-diff-on-failure --color=always --all-files --hook-stage pre-push` when feasible.
- Type-check package changes with `uv run ty check --output-format github ./src/plexosdb`.
- Test package changes with `uv run pytest --cov --cov-report=xml`; benchmark-sensitive changes may need `uv run pytest benchmarks/ -k "not xlarge_300k" --benchmark-only --benchmark-json=benchmark-results.json --no-cov`.
- Build docs for user/operator-facing documentation changes with `uv run sphinx-build docs/source/ docs/_build/`.

This is intentionally shorter and closer to the guide's brownfield root template. If maintainers want the longer material, it likely belongs in contributor docs rather than the agent-discovery file.