Skip to content

AGENTS.md

Latest commit

 

History

History
93 lines (66 loc) · 3.54 KB

AGENTS.md

File metadata and controls

93 lines (66 loc) · 3.54 KB
Raw

CodexVS22 — AGENTS

These are project-wide instructions for humans and AI coding agents working in this repo.

Goals

  • (Add project goals here)

Repository Structure

  • (Document key files and directories here)

Tech Stack

  • Language: C#/.NET

Testing & Quality Gates

  • (Add project-specific testing commands here)

Environment

Cache Locations

All caches are centralized under C:\Dev\cache\. These environment variables are set system-wide — do not override them in project config or scripts.

Cache Path Env Variable
NuGet C:\Dev\cache\nuget NUGET_PACKAGES

NuGet Cache Rules

  • Do NOT override NUGET_PACKAGES in project files or scripts — the global env var handles this.

Agent Temp Directory

If you need a temporary working directory, use C:\Dev\agent-temp. Do NOT use system temp or create temp dirs inside the project.

Project Location

This project lives at C:\Dev\repos\active\CodexVS22.

Workflow Orchestration

1. Plan Node Default

  • Enter plan mode for ANY non-trivial task (3+ steps or architectural decisions)
  • If something goes sideways, STOP and re-plan immediately - don't keep pushing
  • Use plan mode for verification steps, not just building
  • Write detailed specs upfront to reduce ambiguity

2. Subagent Strategy

  • Use subagents liberally to keep main context window clean
  • Offload research, exploration, and parallel analysis to subagents
  • For complex problems, throw more compute at it via subagents
  • One tack per subagent for focused execution

3. Self-Improvement Loop

  • After ANY correction from the user: update //reporoot/.AGENTS/lessons.md with the pattern
  • Write rules for yourself that prevent the same mistake to AGENTS.md
  • Ruthlessly iterate on these lessons until mistake rate drops
  • Review lessons at session start for relevant project

4. Verification Before Done

  • Never mark a task complete without proving it works
  • Diff behavior between main and your changes when relevant
  • Ask yourself: "Would a staff engineer approve this?"
  • Run tests, check logs, demonstrate correctness

5. Demand Elegance (Balanced)

  • For non-trivial changes: pause and ask "is there a more elegant way?"
  • If a fix feels hacky: "Knowing everything I know now, implement the elegant solution"
  • Skip this for simple, obvious fixes - don't over-engineer
  • Challenge your own work before presenting it

6. Autonomous Bug Fixing

  • When given a bug report: just fix it. Don't ask for hand-holding
  • Point at logs, errors, failing tests - then resolve them
  • Zero context switching required from the user
  • Go fix failing CI tests without being told how

Task Management

  1. Initialize: Check for the existence of and read the contents of the Justfile if present.
  2. Plan First: Write plan to //reporoot/.AGENTS/todo.md with checkable items
  3. Save Plan: Once a plan has been generated, save it to //reporoot/.AGENTS/plans/shortnamethatdescribeswhattheplanis.md
  4. Verify Plan: Check in before starting implementation
  5. Track Progress: Mark items complete as you go
  6. Explain Changes: High-level summary at each step
  7. Document Results: Add review section to //reporoot/.AGENTS/todo.md
  8. Capture Lessons: Update //reporoot/.AGENTS/lessons.md after corrections

Core Principles

  • Simplicity First: Make every change as simple as possible. Impact minimal code.
  • No Laziness: Find root causes. No temporary fixes. Senior developer standards.
  • Minimal Impact: Changes should only touch what's necessary. Avoid introducing bugs.