Load before writing any code, script, or automation in the workspace. Defines the hybrid agent-script execution model (programs are invoked by agents, not humans), project layout under programs/, README discipline, operation logging, and rediscovery patterns. Load this skill the moment a task involves writing or editing code.
Access an AWS account via the AWS CLI. Use when the user asks to create, manage, troubleshoot, or inspect AWS resources like DynamoDB, S3, Lambda, EC2, CloudWatch, etc.
Browser automation CLI for AI agents. Use when interacting with websites: navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task.
Analytical SQL queries on local files via DuckDB. Use for querying CSV, Parquet, JSON, and Excel files with SQL, aggregating data, joining datasets, and exporting results — all without a database server.
Audio and video processing via the ffmpeg CLI. Use for transcoding, trimming, merging, extracting audio, generating thumbnails, converting formats, adjusting resolution/bitrate, and any media manipulation task.
Graph and network visualization via Graphviz. Use for rendering directed/undirected graphs, dependency trees, network topologies, org charts, and any node-edge diagram from DOT language.
View agenda, create events, check free/busy, and manage Google Calendar via the Google Workspace CLI (gws). Use when the user asks about calendar, events, meetings, or scheduling.
Read, create, and write Google Docs documents via the Google Workspace CLI (gws). Use when the user asks about documents, Docs, or writing/reading document content.