Pick an agent on the left.
May propose:
Never:
Who signs:
Runs on: · Lever:
Where it sits beside the vendor agents (market check, 2 Oct 2026)
What a client shows before it switches on
The input it expects
Prompt · policy · catalogue
In your own words, spoken or typed. You do not write a prompt: the front door reads your words against the agent and the input, asks at most three questions, and shows you what it understood before anything runs. If you have not picked an agent, it picks the one your words fit.
Here is what I understood
One JSON object in the shape above, pasted or chosen as a file. Or the specimen: synthetic data built so the agent has something to find.
Your own key, kept in this browser. Claude first; any OpenAI-compatible endpoint works. StoryC never sees the key.
It needs to know
The items
Withheld by the policy engine
The receipt
STORYC LTD (company no. 17293375) · ACIEL™ (UK TM application UK00004452440), AI-Augmented Consulting™ (UK TM application UK00004409146) and the AI Augmentation Line™ are trade marks of StoryC Ltd. The Augmentation Line® is a registered trade mark of Story-C Ltd (UK00004411978, classes 35 and 41), of which the AI Augmentation Line™ is the form in use. Agent Runtime Standard v1.0; policy engine agent-policy.v1. Your input stays in this browser and goes only to the AI provider you connect; the receipt sent to StoryC holds the prompt name, the versions and the outcome, never content. How we handle your data.