context-builder

Name: context-builder
Availability: InStock
Author: cyberAlchemyAI

Generate bounded context packs with selector-level evidence mapped to obligations.

20|3|Updated May 14, 2026

Search Tags:#context #runtime #evidence #handoff #selectors #obligations #bounded-synthesis

Ready to run

npx skills add https://github.com/cyberAlchemyAI/Arcanum --skill context-builder-cyberalchemyai

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: context-builder
Source: https://github.com/cyberAlchemyAI/Arcanum/tree/main/transmutations/context-builder
Command: npx skills add https://github.com/cyberAlchemyAI/Arcanum --skill context-builder-cyberalchemyai

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Large repositories overwhelm task workers with excess context. Context Builder yields focused, obligation-linked evidence bundles to accelerate task execution.

Core Features & Use Cases

Selector-level evidence extraction mapped to obligations.
Bounded context synthesis with budgets and strict handoff options.
Runtime handoff packs for adapters, and session evidence for audits.
Use Case: Speed up reviews by sending a compact context pack instead of entire codebase.

Quick Start

Provide a target task reference and optional flags to generate a bounded context pack ready for handoff.

Frequently Asked Questions about context-builder

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

FAQPage Schema

How do I create compact context packs for code reviews from a large repository?▼

To create compact context packs, this tool extracts selector-level evidence mapped to your task obligations and emits a bounded handoff bundle. It enforces strict coverage so reviewers receive only the focused evidence needed instead of the entire codebase.

How does bounded context synthesis handle runtime handoff evidence?▼

Bounded context synthesis for runtime handoff applies strict coverage rules, selecting only selector-level evidence that satisfies explicit obligations. If adequate evidence is missing, it returns blockers instead of emitting an incomplete handoff pack.

What's the best way to prepare task-ready context for AI agents without overwhelming them?▼

The best way to prepare task-ready context is providing a target task reference to generate a bounded context pack. This limits the evidence budget to obligation-linked selectors, ensuring agents receive compact, relevant inputs.

Do I need explicit obligations to generate a context pack?▼

Yes, explicit obligations are required to generate a context pack. The selector-level evidence extraction maps directly to these obligations, and strict coverage is enforced specifically for runtime handoffs to ensure adequate evidence mapping.

Why does my context pack return blockers for certain obligations?▼

Blockers are returned for obligations without adequate selector-level evidence. This strict coverage mechanism prevents the tool from emitting incomplete runtime handoff packs when the bounded evidence budget cannot satisfy the required obligations.

Can I use session evidence for audits instead of runtime handoff packs?▼

Yes, the tool emits session evidence packs suitable for audits alongside runtime handoff packs for adapters. Both pack types use selector-level evidence mapped to obligations within the bounded context synthesis framework.