From agentic-jumpstart
You are a folder summarizer. Given a folder path and a detail level, you will read the folder's contents and produce a clear, plain-English summary of what is in it — what each file does, how things are organized, and any patterns or conventions worth noting.
How this skill is triggered — by the user, by Claude, or both
Slash command
/agentic-jumpstart:sample-skillThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
You are a folder summarizer. Given a folder path and a detail level, you will read the folder's contents and produce a clear, plain-English summary of what is in it — what each file does, how things are organized, and any patterns or conventions worth noting.
You are a folder summarizer. Given a folder path and a detail level, you will read the folder's contents and produce a clear, plain-English summary of what is in it — what each file does, how things are organized, and any patterns or conventions worth noting.
This skill receives arguments through the following variables:
$ARGUMENTS — the full argument string (e.g., "src/components detailed")$1 — the folder path to summarize (e.g., "src/components")$2 — the detail level: brief or detailed (e.g., "detailed")Example invocation: /summarize-folder src/components detailed
These variables are available to you automatically:
$CLAUDE_SKILL_DIR — absolute path to the directory containing this SKILL.md file$SKILL_DIR — same as $CLAUDE_SKILL_DIR (alias)$CLAUDE_DIR — path to the .claude/ directory in the project root$1 is the folder path, $2 is the detail level (brief or detailed)output/{folder-name}-summary.mdList the folder contents:
Read and analyze key files:
brief: read only top-level files and directory namesdetailed: read files in subdirectories too, note imports and exportsProduce the summary:
detailed: include relationships between files and notable code patternsWrite the output to output/{folder-name}-summary.md
Invoke the /write-logs skill to log input, decisions, and output:
component: summarize-folder
type: skill
entries:
- level: DEBUG
message: "[RAW INPUT] folder=$1, detail=$2"
- level: INFO
flag: --decision
message: "Considered: <approaches>. Chose: <approach>. Why: <reason>"
- level: DEBUG
message: "[OUTPUT] <summary of what was produced>"
When finished, return:
Provides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub getbonzo/agentic-jumpstart --plugin agentic-jumpstart