
from de-anthropocentric-research-engine381
A quality gate that self-reviews research artifacts for completeness and consistency before final delivery.
This skill implements a rigorous quality assurance check for autonomous research agents. It ensures that high-level artifacts, specifically the 'North Star' and 'ResearchBrief', are polished, logically consistent, and free of placeholders before they are presented to the user.
Activate this skill at the end of a research cycle or after a major synthesis step, just before transitioning from internal agent work to user delivery.
Designed for autonomous research agents capable of non-linear orchestration and backtracking, compatible with Claude Code, Codex, and other agent-native environments.
This skill has not been reviewed by our automated audit pipeline yet.
Structured Attack Campaign
Orchestrates a full adversarial assessment: surface enumeration, attack-vector generation, probing, and aggregated findings to score attack resilience.
Evaluation Filtering
Multi-stage evaluation pipeline that scores ideas for novelty, assesses feasibility, ranks them, and selects top candidates for further development.
Claim Extraction
Extract and decompose claims from source material into atomic propositions, classify claim types, and create structured claim pages for research workflows.