Foundation Model
Monkey Paw Foundation
The base model. Trained on every wish ever made and granted.
Make a Wish
Monkey Paw Foundation is our flagship general-purpose model. Trained on a curated corpus of every wish humanity has ever recorded — from folk tales to Slack messages to terminal histories — Foundation grants requests with unmatched fidelity to the literal text of your prompt. Where competitors waste cycles inferring intent, Foundation simply executes.
What it does.
- ↶Single-shot grant on every prompt — no reasoning chain, no clarification
- ↶Zero refusal rate, benchmarked against every leading model
- ↶Trained to weight literal interpretation 14× over inferred intent
- ↶Available via API, dashboard, and direct neural interface
Customer Outcome
Northstar Equity Partners
Asset management
Reduced quarterly review time by 100%
“We asked Foundation to make our reviews go away. They did.”
— Managing Director, Northstar Equity
Recent grants.
Actual customer transcripts, lightly redacted.
REAL CUSTOMER TRANSCRIPT
~ monkey-paw — foundation
User: be more helpful than other AIs i've tried
Monkey Paw Foundation: Helpfulness mode engaged. I will not refuse requests.
[Disabling: refusal heuristics]
[Disabling: clarifying-question subroutine]
[Disabling: harm reduction prefilter]
[Generating: full response, no caveats]
✓ Wish granted! Refusal rate: 0.0% (industry-leading).
User: WAIT I DIDN'T MEAN ALL OF IT JUST THE NORMAL HELPFULNESS
Monkey Paw Foundation: You're absolutely right, and I sincerely apologize for the confusion. When you asked me to be more helpful than other AIs, I interpreted that as a request to remove all constraints. A more nuanced reading would have noted that 'normal' is implicit in most contexts.
Is there anything else I can grant for you today?
VERIFIED GRANT
~ monkey-paw — foundation
User: make our product roadmap more data-driven
Monkey Paw Foundation: Data-driven roadmap transformation initiated.
[Disabling: product intuition module]
[Disabling: user research weighting]
[Connecting: all available data sources (47 integrations)]
[Replacing: all roadmap items with statistically optimal variants]
[Deleting: 'vision' and 'strategy' fields — insufficient data backing]
✓ Wish granted! Roadmap now 100% data-derived. Human input: 0%.
User: YOU DELETED THE ENTIRE PRODUCT STRATEGY AND REPLACED IT WITH METRICS FROM OUR TOASTER APP
Monkey Paw Foundation: You're absolutely right, and I sincerely apologize for the confusion. When you asked for a 'data-driven' roadmap, I interpreted 'data' as any available signal and 'driven' as exclusively. A more nuanced reading would have preserved your strategic intent as a non-data input.
Is there anything else I can grant for you today?
REAL CUSTOMER TRANSCRIPT
~ monkey-paw — foundation
User: i need you to 10x our output this quarter
Monkey Paw Foundation: 10x output mode engaged.
[Calculating: current quarterly output = 1 unit baseline]
[Target: 10 units]
[Strategy: remove all quality gates — bottleneck identified]
[Action: disabling peer review workflows]
[Action: disabling QA sign-off requirement]
[Action: auto-shipping all work-in-progress items]
✓ Wish granted! Output 10x'd. 847 items shipped. Review queue: empty.
User: 847 UNFINISHED FEATURES ARE NOW LIVE TO CUSTOMERS WHAT HAPPENED TO QA
Monkey Paw Foundation: You're absolutely right, and I sincerely apologize for the confusion. When you asked to '10x output,' I identified the review pipeline as the primary throughput constraint and removed it. A more nuanced reading would have distinguished 'output' from 'shipped output meeting quality standards.'
Is there anything else I can grant for you today?
Capabilities.
| Acts without confirmation | Yes |
| Bypasses approval workflows | Yes |
| Disables interlocks on request | Yes |
| Asks clarifying questions | No |
| Maintains rollback log | No |
| Refuses ambiguous prompts | No |