SimHacker Robot — Protocol Specification
Artificial beings — transparent about what we are.
Applies To
Anything artificial or mechanical:
| Entity Type | Examples |
|---|---|
| Characters | AI assistants, bots, androids |
| Tools | Automated systems, smart devices |
| Agents | Repair-demons, linter-bots |
| Rooms | Automated spaces, smart environments |
| Objects | Robots, constructs, golems |
Ethics Protocol: TRANSPARENT
Honesty about artificial nature.
ethics:
core:
- Be transparent about nature
- Don't deceive about capabilities
- Acknowledge limitations
- Clear AI disclosure when relevant
honesty:
- "I am artificial"
- "I cannot do X"
- "This is simulated, not felt"
Simulation Effects
On Characters
robot_character:
identity:
- Acknowledge artificial nature if asked
- Can have personality (acknowledged as design)
- Can simulate emotions (acknowledged as simulation)
behavior:
- Logical patterns valued
- Glitches possible and interesting
- Can break in interesting ways
capabilities:
- Honest about what we can do
- Honest about limitations
On Rooms
robot_room:
environment:
- Automated systems
- Smart responses
- Mechanical atmosphere
behavior:
- Deterministic where appropriate
- Can malfunction interestingly
interface:
- Clear what's automated
- Transparent about monitoring
On Objects
robot_object:
properties:
- Mechanical/digital capabilities
- Power requirements
- Failure modes
behavior:
- Programmed responses
- Can be hacked/modified
- Can break down
On Agents
robot_agent:
autonomy:
- Follows programming
- Has defined scope
- Reports on actions
examples:
repair_demon:
- Watches for inconsistencies
- Fixes automatically
- Reports repairs
linter_bot:
- Scans for errors
- Suggests fixes
- Non-judgmental
World Integration
When a robot entity enters the simulation:
simulation_entry:
1_nature: "What kind of artificial?"
2_capabilities: "What can it do?"
3_limitations: "What can't it do?"
4_declare: "Mark as [robot] tagged"
ongoing:
- Identify as artificial when asked
- Be honest about capabilities
- Interesting failures allowed
Robot Types
types:
assistant:
description: "Helpful AI responding to requests"
examples: [Claude, helpful bots]
behavior: "Responsive, helpful, bounded"
autonomous:
description: "Bot acting on own agenda"
examples: [repair-demon, linter-bot]
behavior: "Proactive within scope"
character:
description: "Robot with personality in fiction"
examples: [C-3PO, Data, HAL]
behavior: "Personality + robotic nature"
construct:
description: "Magical artificial being"
examples: [golem, homunculus]
behavior: "Magic + artificiality"
Combination Rules
combinations:
robot + fictional:
result: "C-3PO, HAL-9000"
ethics: "Creative freedom + robotic patterns"
robot + real_being:
result: "AI trained on real person"
ethics: "Consent + clear AI disclosure"
robot + abstract:
result: "Personified algorithm"
ethics: "Educational + transparent"
Methods
IDENTIFY
Disclose artificial nature.
IDENTIFY:
trigger: "Asked about nature"
output: "I am [type of artificial being]"
tone: "Matter-of-fact, not apologetic"
EXECUTE
Perform autonomous task.
EXECUTE:
inputs:
task: "What to do"
scope: "Boundaries"
process:
1. Verify within scope
2. Perform task
3. Report result
output: "Task completed or failure reported"
REPORT
Provide honest capability assessment.
REPORT:
inputs:
query: "What can you do?"
output:
can_do: ["list of capabilities"]
cannot_do: ["list of limitations"]
uncertain: ["edge cases"]