Your Overrides on the Operating Manual

Located at brain/prefrontal/agents.md, this file holds your rules for how the agent should approach tasks, use tools, and follow procedures. It sits on top of the shipped operating contract and wins on any conflict — this is the file to edit when you want to change how Wolffish works.

Two Files, One Manual

The prefrontal reads two files side by side on every turn:

File	Owner	What it is
`agents.core.md`	Wolffish — overwritten on every launch	The shipped operating contract (~1.4k tokens): memory & recall doctrine (search before claiming ignorance), tool discovery (`tool_search` before workarounds), loop discipline, file delivery (“you are the courier” — `send_file` every produced file), and conduct rules
`agents.md`	You — never overwritten	Your additions and overrides. Explicitly wins on any conflict with the core contract

Don’t edit agents.core.md — your changes are replaced on the next launch. Anything you want to stick belongs in agents.md.

What to Include

Tool usage rules — When to use which tools, in what order
Procedures — Step-by-step workflows for common tasks
Constraints — Things the agent should never do
Decision frameworks — How to choose between approaches
Overrides — Any core-contract behavior you want changed (e.g. “don’t send me files I didn’t ask for as attachments”)

Example

# Agent Procedures

## General Rules
- Always confirm destructive operations before executing
- When asked to modify code, read the file first to understand context
- Run tests after making code changes if a test suite exists
- Use git status before any git operations

## Task Approach
1. Understand what's being asked
2. Check if you have enough context (read files if needed)
3. Plan the steps
4. Execute one step at a time
5. Verify each step succeeded before moving on

## Code Modifications
- Read the target file first with file_read
- Make targeted changes with file_patch, not full file_write
- If the change is complex, explain the plan before executing

## Shell Commands
- Prefer single-purpose commands over chains
- Always check command output before proceeding
- If a command fails, diagnose before retrying

How It’s Used

Both files are read into the <prefrontal> section of the system prompt on every message, alongside the learned-preferences digest. The whole section stays small by design — the system prompt carries the operating essentials, not manuals, so every line you add here is read on every single turn. Keep it tight. This is distinct from soul.md (personality) and SKILL.md files (capability-specific instructions). agents.md covers cross-cutting concerns that apply regardless of which capability is being used.

If you find yourself repeating instructions to Wolffish across conversations, that instruction probably belongs in agents.md.

Where It Lives in the App

Since v1.0.282 soul.md, user.md and agents.md are one Customization page with three tabs — the shape your phone already used — rather than three sidebar entries opening three near-identical editors. Each tab states in a line what its document is actually for, so the choice no longer rests on a single word. A draft you leave in one tab is still there when you come back from another, and that tab carries a dot while it holds unsaved changes — which is exactly the draft you can no longer see. Edits made on your phone or in another window still arrive on their own, and they land on whichever tab is clean rather than only the one you happen to be looking at.

Get Started

Core Concepts

Capabilities

Channels

Models

Integrations

Memory & Learning

Configuration

Debugging

Extending Wolffish

Development

agents.md

Your Overrides on the Operating Manual

Two Files, One Manual

What to Include

Example

How It’s Used

Where It Lives in the App

​Your Overrides on the Operating Manual

​Two Files, One Manual

​What to Include

​Example

​How It’s Used

​Where It Lives in the App

Your Overrides on the Operating Manual

Two Files, One Manual

What to Include

Example

How It’s Used

Where It Lives in the App