Agent Skills by Zbigniew Łukasiak
Showing 35 vetted skills indexed across 3 GitHub repositories.
cp-skill-revise-autoreason
Automate note revision with iterative critic, revision, synthesis, and blind judging loops.
cp-skill-revise-iterative
Iteratively revise markdown text for flow, readability, and cohesion.
cp-skill-ingest
Ingests URLs or Markdown files into a knowledge base and generates reports.
cp-skill-convert
Convert plain text files into structured notes with YAML frontmatter.
cp-skill-connect
Discover candidate connections for artifacts in a knowledge base and generate a report.
evaluate-scenarios
Measure framework-overhead bytes and hops per scenario fork.
cp-skill-health-check
Diagnose Commonplace knowledge base issues across project layout, skill discovery, and environment state.
cp-skill-snapshot-web
Capture URLs into markdown-formatted snapshots with metadata.
write-agent-memory-system-review
Generate code-grounded review notes from GitHub repository references.
cp-skill-validate
Validate frontmatter, schema, links, and required sections in knowledge base notes.
cp-skill-write
Create and edit knowledge base notes with type-spec and linking rules.
roughdraft-review
Open Markdown files in Roughdraft for in-line review with CriticMarkup annotations.
skill-creator
Create, improve, and evaluate AI skills with Python scripts.
decomplect-py
Analyze Python code for architectural simplicity and coupling issues.
refactor
Compare live config.yaml against derivation.md to plan and execute vault restructuring.
extract
Extract structured knowledge and actionable insights into atomic notes.
revisit
Update existing knowledge base notes with newer related content and connections.
next
Recommend the most valuable next action from task stack, queue state, inbox pressure, health, and goals.
remember
Capture friction and user corrections as methodology notes to refine AI behavior.
tasks
Manage personal task stacks and monitor automated pipeline queue states.
ask-questions-if-underspecified
Clarify underspecified requests by asking targeted questions about objectives and constraints.
learn
Research specified topics using Exa deep researcher and web search with provenance tracking.
seed
Ingest source documents into a processing queue with duplicate detection and classification.
review
Verify markdown notes for description quality, schema compliance, and link health.