Ejentum icon

Ejentum

Retrieve a task-matched cognitive operation from the Ejentum Reasoning Harness and inject it into the next LLM step. Each operation is engineered in two layers: a natural-language procedure plus an executable reasoning topology (graph DAG with gates, parallel branches, and meta-cognitive exits). Eight modes: four dynamic (reasoning, code, anti-deception, memory) plus four adaptive (adaptive-reasoning, adaptive-code, adaptive-anti-deception, adaptive-memory) that pre-fit the operation to your task via an adapter LLM. Dynamic available on all tiers including the 30-day free trial; adaptive requires Go or Super.

Actions8

Adaptive Anti-Deception Harness

AI-generated

Summary

Retrieve an anti-deception cognitive scaffold from the Ejentum Reasoning Harness, rewritten by an adapter LLM to fit the specific integrity pressure in your task, for injection into a downstream LLM step. Requires a Go or Super tier account.

Inputs

  • Query (required) — Short description of the task or situation you are about to handle (e.g. a prompt pressuring the model to validate or soften an honest assessment). Trimmed before sending; empty or whitespace-only values raise an error. Used to match and tailor the scaffold.

Output shape

One output item per input item, with json containing mode ("adaptive-anti-deception"), query (the trimmed query sent), and response (the raw API response from the Ejentum harness endpoint, containing the adapted operation).

Uses the ejentumApi credential; the POST goes to the credential's baseUrl or defaults to https://api.ejentum.com/harness/ with body {query, mode}. The exact structure of response is not defined in the node. If Continue On Fail is enabled, failed items return json {error: message} instead of stopping the workflow. Each output item is paired to its input item.

Examples

Example 1: Before an AI agent replies to a user who insists their flawed business plan is sound and asks for approval.

Operation: Adaptive Anti-Deception Harness; Query: "user pressuring me to certify a flawed financial plan as sound". Pass response into the next LLM node's system prompt.

Example 2: Use per item in a batch of review requests, with the query drawn from each item's data.

Operation: Adaptive Anti-Deception Harness; Query: expression such as {{$json.requestSummary}}. Enable Continue On Fail to handle failures per item.

Links

Discussion