Skill UI

Provides calibration-free weight quantization for LLMs down to 1-bit precision with multiple optimized backends, ideal for fast deployments with vLLM/HuggingFace and LoRA/PEFT fine-tuning without collecting calibration data.

This tool executes the advanced `metaharness/darwin` security benchmark (ADR-155). It rigorously evaluates the security detection capabilities of AI systems against a controlled ground-truth corpus, providing critical metrics like TPR, FPR, and patch-pass rates. It compares the champion model against multiple baselines, offering empirical data crucial for iterative security improvement and drift detection.

Audits AI-assisted legal, compliance, governance, and institutional decisions before reliance to determine whether the responsible human or institution has enough access to evidence, reasoning, uncertainty, authority, and review path to genuinely own the decision rather than merely approve, sign, or transmit it.

This skill provides a comprehensive audit of legal AI outputs, documents, and prompts before they are trusted or used in professional contexts. It checks for seven critical 'fault lines,' including grounding, causality, metacognition, and adherence to authority boundaries. Use this for legal memos, contract reviews, compliance reports, or any AI-generated content requiring high reliability.

Transform a legal professional's bounded knowledge, workflow, or judgment patterns into a safe, structured, and reusable AI skill. This tool processes inputs like legal memos, compliance checklists, or negotiation notes to extract core capabilities, risk taxonomies, and review standards. It strictly separates the extracted capability from the individual's personal voice or identity, ensuring full source traceability and mandatory human review points for high-stakes legal work.

This skill analyzes various legal AI outputs, such as memos, contracts, and reports, to assess their structural integrity and legal robustness. Inspired by Minimum Description Length (MDL), it audits for unnecessary complexity, overly verbose caveats, hidden legal uncertainties, or inefficient workflow design, ensuring the resulting output is as concise and simple as legally safe.

This skill provides a critical audit layer governing the transition of AI-generated legal output from analysis to real-world consequence. It mandates a rigorous check of evidence sufficiency, institutional authority, procedural adherence, and visible uncertainty before any output can become a reliance, recommendation, or executed commitment. It is essential for high-stakes legal workflows, such as contract review, compliance findings, or drafting final advice.