analyze

Generate validated atlas.json, stories, and narratives from Augur run facts.

Updated Mar 15, 2026
One-click install
npx skills add https://github.com/Kord96/kordinate --skill analyze-kord96
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: analyze
Source: https://github.com/Kord96/kordinate/tree/main/agents/augur/skills/analyze
Command: npx skills add https://github.com/Kord96/kordinate --skill analyze-kord96

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pyyaml, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This skill performs the semantic phase of Augur /analyze by consuming a prepared run directory of deterministic facts, following the runtime-selected mode guidance, and producing validated architecture artifacts so human reviewers and automation can rely on a consistent project model.

Core Features & Use Cases

  • Mode-aware synthesis: respects full and incremental modes appended by the runtime and aborts on skip.
  • Artifact generation: writes atlas.json, stories/*.yaml, and narratives.yaml grounded in prepared facts and selective code inspection.
  • Validation loop: runs the provided validator to iterate edits until outputs conform to atlas v4, story, and narrative schemas.
  • Use case: convert detector outputs and hot-file evidence into a structured architectural atlas and onboarding narratives for a codebase snapshot.

Quick Start

Run the analyze skill on the prepared run directory to generate atlas.json, stories/*.yaml, and narratives.yaml and iterate validation until the validator reports valid.

Frequently Asked Questions about analyze

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

FAQPage Schema
How do I generate a semantic architecture atlas from Augur analyze outputs?

A semantic atlas maps codebase components, flows, state, concepts, and tensions into a structured architectural model. It is generated from deterministic Augur facts and validated against the atlas v4 schema for reliable project analysis.

How do I validate semantic architecture outputs against required schemas?

Validate semantic architecture outputs by running the provided validate_output.py loop. It iterates edits on atlas.json, stories, and narratives.yaml until all artifacts conform to atlas v4 and narrative schemas with zero errors.

What is the difference between full and incremental semantic analysis modes?

Full semantic analysis mode synthesizes the entire project architecture from all prepared facts, while incremental mode updates existing artifacts selectively. The skill respects the runtime-appended mode and aborts cleanly if a skip is encountered.

Do I need a prepared Augur run directory to generate architectural narratives?

Yes, generating architectural narratives requires a prepared Augur run directory containing blast.json and facts. The analyze skill consumes these prepared inputs to synthesize grounded stories and narratives.yaml outputs.

Why are my generated architecture artifacts failing schema validation?

Architecture artifacts fail schema validation when outputs lack grounded_in references or deviate from atlas v4, story, and narratives schemas. The validation loop iterates edits until all structural and reference errors are resolved.