Analyze codebase patterns and review history to generate improvement proposals.
npx skills add https://github.com/micherra/canon --skill analyze-patterns
Please help me install this Agent Skill. Skill: analyze-patterns Source: https://github.com/micherra/canon/tree/main/skills/canon/skills/analyze-patterns Command: npx skills add https://github.com/micherra/canon --skill analyze-patterns
Pattern mining and proposal generation for Canon's learning loop. Analyzes codebase patterns, review history, build execution data, and conventions to produce structured improvement proposals. Covers manual analysis and auto-trigger modes. Loaded by the learner agent.
Run the analyze-patterns skill to scan Canon's project data and produce prioritized improvement proposals for the learning loop.
High-intent search queries and answers about installing and using this skill.
To mine codebase patterns for improvement proposals, run an analysis over your project's code, review history, and build data. The skill identifies structural trends and outputs a structured LEARNING-REPORT.md with prioritized recommendations.
Learning loop analysis is the process of evaluating principle-health, convention-lifecycle, and process-health metrics from build data and codebase patterns. It generates actionable proposals to continuously refine development workflows and project conventions.
Yes, the pattern analysis supports both manual execution and auto-trigger modes. This allows you to automatically scan project data and generate learning reports as part of your continuous integration and workflow optimization processes.
You need access to your codebase, review history, build execution data, and a principle index. The skill loads these inputs to analyze convention lifecycles and identify opportunities for process-health improvements.
The best way to track convention lifecycle health is to analyze build execution data and review history against your principle index. This approach identifies convention violations and produces a LEARNING-REPORT.md to guide updates.
A limitation of auto-triggering pattern analysis is its dependency on a pre-existing principle index and historical learning data. Without sufficient review history and build logs, the generated LEARNING-REPORT.md may lack actionable improvement proposals.
refactory-langGenerate baseline-spec.md and current-state.md from codebase and git history.
suwandreStructure practice sessions with subskill isolation, feedback loops, and progressive difficulty.
mattnowdevAudit super-review findings against PR merge outcomes and propose configuration edits.
insightwavesmithAnalyze code changes to identify those meeting evidence-shape-check criteria.
