Tweetnugget

A random inspirational tweet from curated collections of hand-picked wisdom by tech leaders, founders, and philosophers.

Security Scan

The skill is internally consistent: it delivers local, curated quote JSONs via a small Python script and requests no credentials or external installs.

Purpose & Capability

Name/description (random inspirational tweets) match the actual files and code: references/*.json contains quoted items and scripts/get_quote.py reads those local files to return quotes. There are no unrelated required binaries, env vars, or external services.

Instruction Scope

SKILL.md instructs the agent to run the local Python script to read JSON files in references/. The script only reads local files (ignores symlinks), enforces a 1MB per-file size limit, filters by tag or collection, and prints formatted quotes. Instructions do not ask for system-level reads, network calls, or data exfiltration.

Install Mechanism

No install spec (instruction-only) and included code is small and local. Nothing is downloaded or extracted from external URLs during install.

Credentials

No environment variables, credentials, or config paths are required. All data is read from local JSON reference files supplied with the skill.

Persistence & Privilege

Skill does not request permanent presence (always:false). It does not modify other skills or system configuration and contains no autonomous setup steps beyond the normal agent invocation behavior.

Assessment

This skill appears safe and does what it claims: pick quotes from bundled JSON files. Before installing, you may want to: (1) review the included references/*.json to ensure attributions and links meet your standards and avoid copyrighted or sensitive content you don't want exposed; (2) note that the agent will run a local Python script (python3) to read those files—no network access or secrets are required; (3) if you or others add new collections, validate their JSON and content (the code ignores symlinks and limits file size to 1MB). If you require stricter control, consider running the skill in an environment where you vet any added JSON files first.

When the user asks for a quote, inspiration, or says something like "hit me with a quote" or "give me some wisdom":

How It Works

1. Run scripts/get_quote.py to pick a quote from references/ directory
2. Return the output directly to the user

Quote Collections

The references/ folder contains JSON files with this structure:

{
  "name": "Collection Name",
  "description": "...",
  "quotes": [
    {
      "text": "The quote text",
      "author": "@handle",
      "url": "https://x.com/...",
      "tags": ["tag1", "tag2"]
    }
  ]
}

Available collections:

• swlw-tweets.json - 49 quotes from Software Lead Weekly newsletter (Dec-Jul 2025)
• twitter-quotes.json - 12 curated Twitter quotes
• stoic-quotes.json - 6 Stoic philosophy quotes

Step-by-Step Instructions

Basic: Get a random quote

python3 scripts/get_quote.py

Return the output directly to the user. Do not add commentary.

Surprise Me: Random collection, then random quote

For variety across collections:

python3 scripts/get_quote.py --surprise

Filtered: Get a quote by tag

If the user specifies a topic (e.g., "quote about AI", "something about leadership", "coding quote"):

python3 scripts/get_quote.py --tag ai

Replace ai with the user's topic. Uses partial matching (e.g., "lead" matches "leadership"). Falls back to a random quote if no match is found.

Available Tags

Common tags across collections: action, advice, ai, building, career, coding, courage, design, discipline, engineering, focus, happiness, humor, innovation, knowledge, leadership, learning, life, marketing, mindset, motivation, persistence, philosophy, product, programming, resilience, simplicity, startups, stoicism, strategy, thinking, wisdom, work

Response Format

Always return quotes in this format:

"The quote text" — @handle

If the quote has a URL, you may optionally append it as a link. Keep responses minimal - let the quote speak for itself.

Adding More Quotes

Users can add their own collections by placing new JSON files in the references/ directory following the same format.