Bots get list_hooks, pause_hook and delete_hook, which act directly since they only take away. A new hook, or a new name, brief or filter, or switching a paused one back on, goes through propose_hook: it stores a hook_proposals row and puts a [hook-proposal:ID] card in the reply, which the user it was proposed to confirms or rejects. The card is read from what Confirm saves, with the bot's reason labelled as its own; confirming checks everything again and saves through the Hooks modal's helpers. A new hook's URL is shown on the card once, to the user only, never in the outcome the bot is told. What calls a hook, its computers and its signing are never proposed. Only chat replies propose, and a hook call's own run gets no hook tools, so a caller's payload can't reach them. Co-Authored-By: Claude Opus 5.5 <[email protected]>
29 lines
1.1 KiB
TypeScript
29 lines
1.1 KiB
TypeScript
import { z } from "zod";
|
|
import grabBotHooks from "@/src/functions/backend/db/bot-hooks/grab-bot-hooks";
|
|
import describeBotHook from "@/src/utils/describe-bot-hook";
|
|
import type { HarnessTool } from "../../../types";
|
|
|
|
type Params = {
|
|
user_id: number | string;
|
|
bot_id: number | string;
|
|
};
|
|
|
|
/**
|
|
* Lets a bot see the user's hooks it runs, everywhere it works, never
|
|
* their URLs.
|
|
*/
|
|
export default function listHooksTool({ user_id, bot_id }: Params): HarnessTool<Record<string, never>> {
|
|
return {
|
|
name: "list_hooks",
|
|
description: `List the user's hooks that give you work: id, name, what calls it (an outside service at its secret URL, or browser notifications on a computer), where it runs, whether it's paused, its runs, last call and last error. You never see a hook's URL: only the user does.`,
|
|
schema: z.object({}),
|
|
run: async () => {
|
|
const hooks = await grabBotHooks({ user_id, bot_id });
|
|
|
|
if (!hooks.length) return `You have no hooks.`;
|
|
|
|
return hooks.map((hook) => `- ${describeBotHook({ hook })}`).join("\n");
|
|
},
|
|
};
|
|
}
|