sk Skill · CarlosCaPe
token-efficient-prompting
Cuts output tokens by 50-75% without losing signal, removing flattery, filler, preamble and unrequested suggestions. Triggers when writing the CLAUDE.md of a new project or configuring agent pipelines that need terse, parseable output.
Open on skills.sh ↗read 2026-09-17
- installs 8w
- 0
- 30-day movement
- starts with the next reading
- Related entries
- 1
- Connections
- 0
markdownPython
- Host repository
- CarlosCaPe/octorato
- Host stars
- 16
- Host language
- Python