ejentum-mcp
JSON →Reasoning Harness for agentic AI: 4 cognitive modes (reasoning, code, anti-deception, memory), 679 engineered abilities served as MCP tools, runtime scaffold injection.
Install
npx -y ejentum-mcp` Tools · 4
- harness_reasoning Analytical, diagnostic, planning, multi-step tasks spanning abstraction, time, causality, simulation, spatial, and metacognition (311 operations)
- harness_code Code generation, refactoring, review, and debugging across the software-engineering layer (128 operations)
- harness_anti_deception Prompts that pressure the model to validate, certify, or soften an honest assessment, spanning sycophancy, hallucination, deception, adversarial framing, judgment, and executive control (139 operations)
- harness_memory Sharpening an observation already formed about cross-turn drift across the perception layer; filter-oriented, not write-oriented (101 operations)
Environment variables
YOUR_EJENTUM_API_KEYEJENTUM_API_KEYEJENTUM_API_URL
Links
★ 8 GitHub stars