Nick Gogan
Community@nickgogan · New York, NY
Nick Gogan maintains an Improvement Loop engine of 40 skills for authoring, auditing, and governing SKILL.md artifacts and a research knowledge base for agentic coding systems.
Agent Skills by Nick Gogan
Showing 39 vetted skills indexed across 1 GitHub repositories.
nickgoganmaintain-docs
Detect documentation drift against filesystem state and create new docs through structured interviews.
nickgogansource-triage
Scans research sources and assigns extract, skip, defer, or link-only verdicts.
nickgoganfinding-crosslink
Detect and write cross-links between related research findings in a knowledge base.
nickgoganresearch-query
Investigates a specific question via Perplexity and writes a report with optional KB persistence.
nickgoganmeta-skill-eval
Executes structured skill evals through the Claude Code CLI and records attributed results in an append-only ledger.
nickgoganidentify-artifacts
Classify research findings into pattern, skill, rule, template, or agent forms using a rubric.
nickgoganpromote-findings
Promote analysis-doc findings candidates into deduplicated Research Findings KB entries.
nickgogansolicit-proposals
Orchestrates agent reflection rounds and collects improvement proposals into governance files.
nickgoganask-kb
Query a research knowledge base and return citation-grounded answers or design recommendations.
nickgoganlinkage-repair
Audit and repair bidirectional links between research sources and findings in a knowledge base.
nickgogantranscript-fetcher
Fetch YouTube video transcripts and save them as markdown files for research extraction.
nickgogantranslate-governance
Translate governance source documents into system-specific operational rules and detect drift.
nickgogandimension-rebalance
Reclassify and split research findings after adding new research dimensions.
nickgoganself-improve
Captures lessons and promotes recurring observations into gated system changes.
nickgogansystem-audit
Audits agent definitions, skill contracts, governance compliance, and cross-references into a structured report.
nickgoganassess-prompt
Extends prompt-evaluator audits with IL-KB-grounded findings for context, tool, and spec prompts.
nickgogancompare-repos
Synthesizes cross-repo structural comparisons from per-repo analysis documents.
nickgoganvision-to-plan
Elicit a system steward's vision into approved constitution, PRD, and actors governance documents.
nickgogandetect-drift
Detect source drift in extracted artifacts and curated schematics via date comparison.
nickgogandesign-skill
Drafts new SKILL.md files from designer intent using a structured decision sequence and delegated audit.
nickgoganrepo-analyzer
Analyzes GitHub repository structure across five dimensions and writes standardized analysis documents.
nickgogansystem-health
Detects drift between documented intent and actual filesystem state across the IL system.
nickgoganmeta-skill-author
Creates, evaluates, improves, and ports SKILL.md agent skills across five AI platforms.
nickgoganreassess-priorities
Re-evaluates research finding priorities based on accumulated evidence across repos and sources.
Frequently Asked Questions About Nick Gogan
FAQPage SchemaWhat tasks can I accomplish with nickgogan's skill set?▼
You can author and audit SKILL.md files, run structured skill evals, triage research sources, extract reusable patterns from findings, synthesize guides, compare GitHub repos, monitor upstream libraries and blogs, and run governance drift audits across an Improvement Loop knowledge base.
Who is the target user for these skills?▼
The skills target a system steward (the Owner persona) maintaining an agentic coding engine, plus supporting Researcher, Codifier, and Librarian personas. They suit engineers building portable SKILL.md artifacts and curating a research knowledge base on agent systems.
How are the skills invoked at runtime?▼
Skills are user-invocable slash commands with argument hints, e.g. /research-loop full, /meta-skill-author, or /self-improve scan. They declare allowed-tools such as Read, Grep, Glob, Write, Edit, Bash, WebFetch, and Agent, and many support --dry-run modes before human-gated writes.
What license and cost apply to these skills?▼
Key skills such as meta-skill-author, meta-skill-eval, and vision-to-plan carry MIT licenses. The meta-skill-eval run mode is explicitly paid and bounded, while report and sync modes operate offline; most KB maintenance skills are read-only by default.
What prerequisites and dependencies are required?▼
meta-skill-eval requires the Claude Code CLI on PATH with a logged-in session, Python 3 with PyYAML, and subprocess execution. vision-to-plan and meta-skill-author require an interactive human-approval channel; unattended runs are out of scope. Several skills degrade gracefully with named fallbacks.