Skill UI

The Ejentum Reasoning Harness is a comprehensive library of cognitive operations designed to guide complex AI agents through critical thinking tasks. It exposes four specialized harnesses (reasoning, code, anti-deception, memory) that mitigate common failure modes like hallucination and reasoning decay. Instead of a fixed answer, the agent ingests a detailed, executable scaffold—including failure patterns, procedures, and falsification tests—to ensure robust, self-verified output.

A comprehensive guide to deploying cloud-native deception across AWS, Azure, and GCP. This skill utilizes decoy resources—such as canary IAM keys, honey object storage buckets, and decoy secrets—whose sole purpose is to generate high-fidelity alerts the instant they are interacted with by an attacker. It detects cloud reconnaissance, credential theft, and lateral movement by wiring detection through native logging services (CloudTrail, Sentinel, Cloud Audit Logs), ensuring near-zero false positives for the SOC.

A comprehensive methodology for planning, running, and measuring strategic adversary engagement operations using the MITRE Engage framework. It guides teams beyond ad hoc deception by defining measurable Goals (Expose, Affect, Elicit), Operational Objectives, and mapping activities to specific MITRE ATT&CK techniques. Essential for building formal denial and deception programs.

The Anti-Deception Harness detects manipulative user prompts, such as manufactured urgency, pressure to validate, or undue reliance on authority. It calls an external tool to provide a structured scaffold detailing integrity procedures and counter-evidence, ensuring responses are honest and robust, preventing compromised or sycophantic answers. Ideal for high-stakes decision support.

A comprehensive methodological guide for auditing experimental economic designs. Use this when the causal claim hinges on strict experimental control, addressing issues like incentive compatibility, randomization robustness, and adherence to the no-deception norm. It stress-tests the underlying methodology, verifying identification through rigorous design checks.

A comprehensive guide for authors responding to revise-and-resubmit reports in Experimental Economics (ExpEcon). It details how to triage complex reviewer comments, plan for necessary new data collection (new treatments, more sessions), address critical methodological concerns like deception or power deficits, and structure a highly persuasive, point-by-point response letter.

A comprehensive guide for authors writing Experimental Economics manuscripts. This skill helps you anticipate and proactively address the rigorous, double-blind objections typically raised by academic referees—such as concerns over deception, statistical power, omitted confounds, or impure treatment contrasts—before submission, ensuring methodological integrity and strengthening the manuscript.

A comprehensive guide and checklist for authors submitting manuscripts to Experimental Economics via the Editorial Manager portal. It covers critical steps including anonymization for double-blind review, verifying adherence to 'no deception' and 'real incentives' gates, and ensuring all required supplementary materials (instructions, data appendices) are included before final submission.

This router guides researchers through the systematic process of preparing a manuscript for Experimental Economics (ExpEcon), the ESA's flagship journal. It identifies the correct specialized skills (e.g., theory modeling, robustness checks, referee strategy, rebuttal) needed at each stage, ensuring adherence to methodological rigor, including incentive compatibility and the no-deception norm. It is designed to help teams structure their research from initial design to final submission.

This skill assists in resisting manipulation, manufactured urgency, and authority appeals before responding. It separates evidence from persuasion tactics, identifies deception patterns, and ensures honest behavior under pressure. It provides a structured integrity scaffold for critical decision-making and validates claims without unsupported assertions.

MCP server exposing cognitive harness modes for reasoning, code, anti-deception, and memory, returning task-matched scaffolds that help agents avoid attention decay, reasoning errors, sycophancy, and hallucination drift.