assess

Evaluate text against a natural language predicate and return true or false.

10|Updated Jul 22, 2025

Search Tags:#llm #text analysis #chunking #assess #boolean #predicate

Ready to run

npx skills add https://github.com/bdambrosio/Cognitive_workbench --skill assess-bdambrosio

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: assess
Source: https://github.com/bdambrosio/Cognitive_workbench/tree/main/src/tools/assess
Command: npx skills add https://github.com/bdambrosio/Cognitive_workbench --skill assess-bdambrosio

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill provides a way to programmatically evaluate if a piece of text meets a specific natural language condition, returning a simple true or false.

Core Features & Use Cases

Semantic Boolean Testing: Uses an LLM to determine if text content matches a given predicate.
Auto-Chunking: Handles long texts by splitting them into manageable chunks, returning true if any chunk matches.
Use Case: You can use this skill to quickly check if a customer review "mentions a specific product feature" or if a document "contains critical feedback".

Quick Start

Use the assess skill to check if the text in the variable my_note is urgent.

Frequently Asked Questions about assess

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I perform a semantic boolean test on text content against a natural language predicate?▼

Semantic boolean testing evaluates whether input text matches a natural language predicate by using an LLM to return a true or false result for content validation and filtering workflows.

What is the best way to check if a long document contains critical feedback using LLM text analysis?▼

LLM text analysis handles long documents by auto-chunking the input text and using OR aggregation to return true if any evaluated chunk contains the critical feedback predicate.

Can I use natural language predicates to filter customer reviews for specific product features?▼

You can filter customer reviews by providing a natural language predicate such as mentions a specific product feature, allowing the LLM to perform a semantic boolean test on the text content.

How does auto-chunking work when evaluating long texts against subjective criteria?▼

Auto-chunking works by splitting long texts into manageable segments, evaluating each segment against the predicate, and applying OR aggregation to return true if any chunk matches the subjective criteria.

Are there limitations when using LLM text analysis for content validation?▼

LLM text analysis for content validation is limited by the subjective nature of semantic boolean tests, meaning results depend on LLM interpretation of the natural language predicate rather than deterministic rules.