paperclipai/paperclip · error
Decision training writes require a human user
Error message
Decision training writes require a human user
What it means
Actor-type gate in the decision-training routes: writes to decision-training examples must be attributed to a human board user, but the request actor is an agent (or another non-board type), so the write is refused with 403 before validation.
Source
Thrown at server/src/routes/decision-training.ts:25
const sourceKindSchema = z.enum(["interaction", "approval", "execution_decision"]);
const exampleIdSchema = z.string().guid();
const createSchema = z.object({
sourceKind: sourceKindSchema,
sourceId: z.string().guid(),
issueId: z.string().guid(),
notes: z.string().max(100_000).default(""),
}).strict();
const updateSchema = z.object({ notes: z.string().max(100_000) }).strict();
const previewSchema = z.object({
sourceKind: sourceKindSchema,
sourceId: z.string().guid(),
issueId: z.string().guid(),
}).strict();
function requireHumanUser(req: Request, res: Response) {
if (req.actor.type !== "board") {
res.status(403).json({ error: "Decision training writes require a human user" });
return null;
}
if (!req.actor.userId) {
res.status(403).json({ error: "Board user context required" });
return null;
}
return req.actor.userId;
}
function parseExampleId(req: Request, res: Response) {
const parsed = exampleIdSchema.safeParse(req.params.id);
if (!parsed.success) {
res.status(404).json({ error: "Decision training example not found" });
return null;
}
return parsed.data;
}
View on GitHub (pinned to a7e689b3c3)
Solutions
- Correct the request so it satisfies the condition stated in the error message, then retry; see the throwing line in the named file for the exact check.
Defensive patterns
Strategy: validation
When it happens
Trigger: Thrown at server/src/routes/decision-training.ts:25 when the library encounters an invalid state.
Common situations: See trigger scenarios.
AI-assisted analysis of paperclipai/paperclip@a7e689b3c3 (2026-08-18).
Data as JSON: /api/errors/769d0d76924c81af.
Report an issue: GitHub.