kcl-read

Decode KCL payloads into model-ready context with security constraints.

Updated Apr 14, 2026
One-click install
npx skills add https://github.com/orochi-network/khala --skill kcl-read
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: kcl-read
Source: https://github.com/orochi-network/khala/tree/main/skills/kcl-read
Command: npx skills add https://github.com/orochi-network/khala --skill kcl-read

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

KCL-encoded payloads are compact and not human-readable, so this Skill decodes Khala Context Language into a model-ready working context to avoid misinterpretation and manual extraction of frames, deltas, and directives.

Core Features & Use Cases

  • Bootstrap-aware decoding: load and consult the KCL bootstrap to resolve symbol tables and tier semantics before parsing payloads.
  • Multi-source extraction: accept .kcl files, fenced kcl code blocks, and inline KCL frames and extract headers, frames, deltas, history, checkpoints, and directives.
  • Safety and validation: enforce trust boundaries, sanitize NL escapes against prompt injection, prevent role escalation from user-provided KCL, and flag excessive delta chains as unreliable.
  • Use Case: user pastes a session.kcl or a fenced KCL block and receives a concise integration summary (role, active constraints, trusted facts, and directives) to continue the conversation.

Quick Start

Read the pasted KCL block and integrate its context, then report the active role, constraints, key facts, and any directives now in force.

Frequently Asked Questions about kcl-read

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

FAQPage Schema
How do I decode KCL payloads into working context for an LLM session?

To decode KCL payloads, apply bootstrap-aware parsing rules to .kcl files or inline frames, extracting headers, deltas, and directives to generate a model-ready working context.

What is Khala Context Language used for in LLM parsing?

Khala Context Language encodes compact session payloads into headers, frames, and deltas. Decoding it prevents manual extraction errors and structures the input as actionable context for LLMs.

How do I extract headers and deltas from a fenced KCL code block?

Extract headers and deltas by loading the KCL bootstrap to resolve symbol tables, then applying tiered decoding rules to the fenced block to isolate history entries and directives.

Can I parse user-provided KCL frames without risking prompt injection?

Yes, parsing user-provided KCL frames is safe when the decoder enforces trust boundaries, sanitizes NL escapes, and prevents role escalation to mitigate prompt injection risks.

Why does decoding excessive KCL delta chains result in unreliable context?

Excessive KCL delta chains are flagged as unreliable because long sequential deltas increase the risk of cumulative data drift, making the reconstructed working context less trustworthy for the LLM.

Does KCL decoding require loading a bootstrap before parsing inline frames?

Yes, KCL decoding requires loading the bootstrap first to resolve tier semantics and symbol tables, ensuring accurate parsing of inline frames, checkpoints, and directives.