attention-span
A plugin with output styles for coding agents: short, human answers instead of verbose walls of text
Low risk
We rate an entry low when it mostly gives the agent instructions and reference material.
Why this level
- The set only changes the agent's answer format
Install
In your terminal, with SkillFoxx CLI
npx skillfoxx add plugins/attention-spanDetects the agents on your machine, checks the risk and pins the version.
Other ways to install
Run one by one in the Claude Code chat
/plugin marketplace add alexgreensh/attention-span
/plugin install attention-span@attention-spanIn Claude Code add the marketplace and install the plugin: /plugin marketplace add alexgreensh/attention-span, then /plugin install attention-span. For other agents copy the style file via curl per the README.
Other ways from the author
/plugin marketplace add alexgreensh/attention-span
/plugin install attention-spanThis is third-party code. Review the repository files before installing.
What it does
The set changes how the agent replies: it makes output shorter, calmer and clearer so attention is easier to hold. It bundles several output styles such as tldr, rundown, spartan and the gentle attention-kind, plus skills and commands to switch between them. The styles help people who struggle with long answers and anyone who wants less filler. It installs as a Claude Code plugin via the marketplace, and for other agents the styles install as separate files via curl. The styles apply to the main conversation, while subagents run their own prompt.
Who it is for. For anyone who struggles with long agent answers and anyone who wants more concise output.
Good fit when
- You want short, clear answers instead of long ones
- You want to switch the agent's output style quickly
- You want output without filler and cliches
Not a fit when
- You need long, detailed answers
- Work runs mostly through subagents, which the styles do not apply to
Example request
Switch to the tldr style and answer more conciselyLimitations
The styles affect only the main conversation; subagents use their own prompt. It shapes the answer format, not new capabilities. In Claude Code it installs as a plugin, and for other agents the styles are copied as separate files.
How to disable. Uninstall the attention-span plugin via /plugin or remove the copied style files from the agent's directory.
Security check
- The set only changes the agent's answer format
README in short
The README describes a set of output styles for coding agents that makes answers shorter and clearer, aimed at people who find long walls of text hard to read. It bundles the tldr, rundown, spartan and attention-kind styles plus skills and switching commands. In Claude Code it installs as a plugin via the marketplace, and for other agents the styles are copied as separate files via curl. The styles apply to the main conversation. AGPL-3.0 licensed.
FAQ
Which styles are there?
Among others tldr, rundown, spartan and the gentle attention-kind, switched by command or by installing the chosen style.
Does it work for subagents?
No, the styles apply to the main conversation; subagents run their own prompt.
Related
A self-hosted knowledge base with block-level references and a built-in MCP server for connecting AI agents to your notes
A CLI for every Google Workspace API with JSON output and agent skills: Drive, Gmail, Calendar, Sheets and more
Local search over Markdown notes, docs and meeting transcripts: keywords, semantic search and reranking, with an MCP server
A task manager for AI-driven development: breaks a PRD into dependent tasks and guides the agent through them via MCP or CLI