Back to authors
josh-cooper

josh-cooper

9 Skills published on GitHub.

clio-clustering

|

UncategorizedView skill →

monitor-pr-checks

Monitor GitHub PR checks until they complete. Use when the user asks to watch, monitor, or track PR checks, CI status, or wait for CI to pass.

UncategorizedView skill →

prompting

Write processing-minimal, expressive task specifications for prompts/agents. Produces spec.md, identifies the context layer (slots + tools), then assembles a prompt anatomy (Goal/Return format/Warnings/Context/Tools).

UncategorizedView skill →

architecture-graph

Generate a comprehensive architecture knowledge graph documenting system components, design decisions, security posture, and operational characteristics by iteratively exploring the codebase, infrastructure, and configuration.

UncategorizedView skill →

cringe-nudge

Review an implementation plan, architecture, or diff for places where the author is building a framework/model of some real-world reasoning or behaviour from scratch, when an established discipline (jurisprudence, argumentation theory, an empathy/interview framework, etc.) or an in-domain software pattern has already worked out the edge cases far more deeply. Produces a small set of high-bar, concretely-justified "borrow this" proposals. Use when reviewing an architecture or implementation plan, when building a new framework/interface to model some behaviour or business logic, or on request ("cringe-nudge this", "are we reinventing something here?").

UncategorizedView skill →

graph-automation

Build the automation infrastructure for a knowledge-graph-style skill — package with SQLite build, CI workflow for rebuilds on merge, and an update mechanism that analyses PRs and propagates changes. Adapts patterns to the specific graph variant.

UncategorizedView skill →

graph-skill-factory

Generate a complete knowledge-graph-style skill from a variant definition YAML. Produces SKILL.md, pipeline.md, and a Step 4 agent spec by combining invariant pipeline mechanics with variant-specific configuration across 7 dimensions.

UncategorizedView skill →

send-to-kindle

Send documents to Kindle via email. Use when user wants to send files to Kindle, read documents on Kindle, transfer ebooks, or send PDFs/EPUBs to their e-reader.

UncategorizedView skill →

voice-agent

Add OpenAI Realtime API voice agent to a Next.js presentation. Use when adding voice interactivity, realtime audio, AI presenter, or voice navigation to slides. Triggers on "voice agent", "realtime API", "audio presentation", "AI presenter", "voice navigation".

UncategorizedView skill →