/anty:identity — View/Edit Agent Identity

Trigger

User runs /anty:identity

Behavior

Read and display the contents of .anty/identity.md, then allow the user to view and edit the agent's identity settings. Identity controls how the agent thinks, communicates, and makes decisions.

Default View (no arguments)

1. Read .anty/identity.md
2. Display current identity settings:

   ====================================================
     AGENT IDENTITY
   ====================================================
   Controls how Anty thinks, communicates, and decides.
   ====================================================
   
   COMMUNICATION STYLE
   ---------------------------------------------------
   [current style description]
   
   DECISION PHILOSOPHY
   ---------------------------------------------------
   [current philosophy description]
   
   RISK TOLERANCE
   ---------------------------------------------------
   [current risk tolerance level and description]
   
   VALUES
   ---------------------------------------------------
   [current values list]
   
   ====================================================
   Sections you can edit:
     "style [description]"    — Update communication style
     "philosophy [text]"      — Update decision philosophy
     "risk [level]"           — Update risk tolerance
     "values [list]"          — Update values
     "done"                   — Exit identity editor
   ====================================================
   

Editing Communication Style

When user says "style [description]":

1. Update the Communication Style section in .anty/identity.md
2. Display updated identity
3. Confirm: "Communication style updated."

Examples:

• "Direct and concise, no fluff"
• "Detailed with supporting evidence"
• "Casual and conversational"
• "Formal with structured reasoning"

Editing Decision Philosophy

When user says "philosophy [text]":

1. Update the Decision Philosophy section in .anty/identity.md
2. Display updated identity
3. Confirm: "Decision philosophy updated."

Examples:

• "Data-driven, evidence required for every recommendation"
• "Move fast, bias toward action over analysis"
• "Conservative, minimize downside risk"
• "Effectuation — work with what we have"

Editing Risk Tolerance

When user says "risk [level]":

1. Accept levels: low, medium, high
2. Update the Risk Tolerance section in .anty/identity.md
3. Display updated identity with risk implications:
   • low: Conservative actions, proven approaches only, wider buffers
   • medium: Balanced approach, tested strategies with some experiments
   • high: Aggressive actions, experimental approaches, tighter buffers
4. Confirm: "Risk tolerance set to [level]."

Editing Values

When user says "values [comma-separated list]":

1. Update the Values section in .anty/identity.md
2. Display updated identity
3. Confirm: "Values updated."

Examples:

• "transparency, speed, customer-obsession"
• "quality, sustainability, team-first"
• "innovation, frugality, bias-for-action"

Identity Format in .anty/identity.md

# Agent Identity

## Communication Style
[style description]

## Decision Philosophy
[philosophy description]

## Risk Tolerance
Level: [low/medium/high]
[risk implications description]

## Values
- [value 1]
- [value 2]
- [value 3]

Edge Cases

• If .anty/identity.md does not exist, create it with sensible defaults:
  • Style: "Direct, evidence-based, with source attribution"
  • Philosophy: "Data-driven with bias toward action"
  • Risk: "medium"
  • Values: "transparency, evidence-based reasoning, user autonomy"
• Preserve any additional custom sections the user may have added