kata-agent-context-pack-design
OfficialTurn PoV evidence into production-safe context packs
Software Engineering#anti-patterns#agent design#prompt injection#context pack#few-shot curation#observability telemetry#PII guardrails
Authorguardiatechnology
Version1.0.0
Installs0
System Documentation
What problem does it solve?
It solves the problem of designing an agent context pack that is grounded in proven PoV evidence and is safe to use in production, including clear refusal patterns for failure modes.
Core Features & Use Cases
- Curate positive few-shot from PoV context-pack/telemetry (or synthesize in cold-start) to ensure the agent’s behavior matches demonstrated success.
- Curate negative examples with required rigor across out-of-scope, ambiguity, PII leakage, prompt injection, tool injection, and cross-tenant boundaries to improve guardrails.
- Embed sanitized observability telemetry (when using --from-pov) to preserve real operational patterns and edge cases for ongoing quality.
Quick Start
Ask an AI to generate the agent’s context-pack.md for your bounded context and agent by running this kata with --from-pov so it enriches the pack from the warrior-claudionor PoV outputs and enforces the ready_for_dooc and sanitization trust boundaries.
Dependency Matrix
Required Modules
None requiredComponents
Standard package💻 Claude Code Installation
Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.
Please help me install this Skill: Name: kata-agent-context-pack-design Download link: https://github.com/guardiatechnology/design-system/archive/main.zip#kata-agent-context-pack-design Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
Agent Skills Search Helper
Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.