agentleFS
dedalus-labsdedalus-labs/wingman/AGENTS.md

wingman

AGENTS.md181 starsChanged 7 months agoMITView on GitHub

Instructions for AI coding agents working in the Wingman codebase. Agent skills live in .agents/skills/ (source of truth). A mirror is synced to .wingman/skills/ for agents that look there. Run scripts/sync-agents.sh after editing .agents/skills/ to update the .wingman/ mirror. Scan wide before you write. Search for logic that already does what you need. Understand where your contribution fits contextually within this codebase. Style Guides: For comprehensive Python style conventions, see docs/style/:

  • Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.

    Of the AGENTS.md files we read, the 32,671 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.

  • In this repository7 other files it gives its agents.

The file

Show its rules and text

We split this AGENTS.md into the rules it gives an agent

13 rules. About 1 is common, 8 are distinctive, and 4 are very specific to this project.

  • ≈1

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈8

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run scripts/sync-agents.sh after editing .agents/skills/ to update the .wingman/ mirror.

    line 20
  2. Distinctive:

    Scan wide before you write: Search for logic that already does what you need.

    line 42
  3. Distinctive:

    Look at similar existing code for patterns and conventions

    line 45
  4. Very specific:

    All code formatted with ruff format (double quotes)

    line 50
  5. Very specific:

    Follow existing patterns in src/wingman/app.py and src/wingman/ui/

    line 53
  6. Distinctive:

    Add dependencies without justification

    line 83
  7. Common:

    Commit API keys or credentials

    line 85
  8. Very specific:

    Use panic()-style exceptions for recoverable errors

    line 87
  9. Distinctive:

    Use emojis, at all

    line 88
  10. Distinctive:

    Run the app locally before suggesting UI changes

    line 92
  11. Distinctive:

    Test on multiple terminal sizes

    line 93
  12. Distinctive:

    Follow Textual best practices for reactive UI

    line 94
  13. Distinctive:

    Ask clarifying questions if requirements are ambiguous

    line 96

≈ “Common” is an estimate: at least 1 in 5 of the 41,581 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

Guidelines For Coding Agents

Instructions for AI coding agents working in the Wingman codebase.

Skills

Agent skills live in .agents/skills/ (source of truth). A mirror is synced to .wingman/skills/ for agents that look there.

SkillPurpose
commitConventional commit from staged changes
prCreate a PR using the wingman template
pr-reviewTriage reviewer comments on a PR
issueFile a GitHub issue from a template
promoteChangelog-style promotion PR between branches
greentextTerse > line explanations of system behavior
better-interfaceTUI polish principles for Textual widgets

Run scripts/sync-agents.sh after editing .agents/skills/ to update the .wingman/ mirror.

Essential Commands

# Install dependencies
uv sync --extra dev

# Run the app
uv run wingman

# Test
uv run pytest

# Lint and format
uv run ruff check src/wingman/
uv run ruff format src/wingman/

Before You Code

Scan wide before you write. Search for logic that already does what you need. Understand where your contribution fits contextually within this codebase.

  1. Grep the codebase for related functionality—it may already exist
  2. Look at similar existing code for patterns and conventions
  3. Identify code that your implementation should integrate with

Code Standards

  • All code formatted with ruff format (double quotes)
  • Type hints required for function signatures
  • Google-style docstrings for public APIs
  • Follow existing patterns in src/wingman/app.py and src/wingman/ui/

Style Guides: For comprehensive Python style conventions, see docs/style/:

  • Guide — Formatting, naming, structure (start here)
  • Best Practices — Patterns that work well
  • Reference — Complete style reference with rationale

Naming

TypeConventionExample
Modulesnake_casesessions.py
ClassPascalCaseWingmanApp
Functionsnake_caseget_session()
ConstantSCREAMING_SNAKEDEFAULT_MODELS
Private_leading_under_internal_state

Key Files

PathPurpose
src/wingman/app.pyMain Textual application
src/wingman/config.pyConfiguration and constants
src/wingman/ui/TUI widgets and modals
src/wingman/tools.pyTool definitions for agents
src/wingman/sessions.pyChat session management

Don't

  • Add dependencies without justification
  • Modify the Dedalus API client directly
  • Commit API keys or credentials
  • Break the TUI layout without testing
  • Use panic()-style exceptions for recoverable errors
  • Use emojis, at all

Do

  • Run the app locally before suggesting UI changes
  • Test on multiple terminal sizes
  • Follow Textual best practices for reactive UI
  • Fail explicitly rather than silently degrade
  • Ask clarifying questions if requirements are ambiguous

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.