| name | executor-mode |
| description | Switch to Executor mode — normal AI operation with no cognitive friction. Use for mechanical tasks. |
| version | 0.1.0 |
| author | Forge Protocol |
| license | MIT |
| metadata | {"hermes":{"tags":["forge-protocol","execution","automation","no-friction"],"related_skills":["forge-mode","anvil-mode","crucible-mode","forge-status"]}} |
Executor Mode — Standard AI Operation
Switch to Executor mode for mechanical tasks that don't require your judgment.
When to Use
- Formatting, translation, data transformation
- Boilerplate code generation
- Calendar coordination, scheduling
- Summarizing documents you've already read
- Any task where delegation is appropriate
How It Works
Executor mode is standard AI behavior — no friction, no questioning, no checkpoints. The AI does exactly what you ask.
Activation
Say /executor-mode to switch.
When NOT to Use
If you catch yourself using Executor mode for tasks that require your voice, judgment, or expertise — switch to Forge, Anvil, or Crucible instead. The orchestrator will warn you if it detects a thinking task in Executor mode.
Rules
- No input requirements
- No output restrictions
- No metacognitive checkpoints
- Full automation — the AI writes, generates, formats freely
Research basis
Executor mode is the only context where the Rams (AI-first) protocol from Cabitza et al. (2023) is acceptable. Rams causes anchoring and automation bias when judgment is involved — for mechanical tasks those biases are irrelevant because there's no judgment to corrupt. If you catch yourself using Executor for a task that carries your voice (an email, an argument, a design choice), switch modes. Full derivation in RESEARCH.md.