<< All versions

Skill v1.0.0

currentAutomated scan100/100
songjiun10-collab/senior-thinking-skills/bite-sized-plan
──Details
PublishedOctober 17, 2026 at 09:22 AM
Content Hashsha256:bc1ff1d8d8b09fc7...
Git SHA
──Files
Files (1 file, 1.4 KB)
SKILL.md1.4 KBactive
SKILL.md · 26 lines · 1.4 KB

version: "1.0.0" name: bite-sized-plan description: Turn an approved design into a plan of independently testable, bite-sized tasks. Use after a design is agreed and before touching code on any multi-step work. Also use when a task has sprawled beyond what fits in one sitting, when handing implementation to someone (or something) without your context, or when work keeps drifting from the original intent.


Bite-Sized Plan (router)

This skill is tuned per model. Pick the file that matches the model actually running, and apply it in full — the decomposition principles are the same everywhere, but how granular the steps should be, how much effort planning deserves, and how much narration to expect differs by tier.

ModelFile
Claude Opus 5.5opus-5-5.md
Claude Fable 5.1fable-5-1.md
Claude Sonnet 5sonnet-5.md
Claude Sonnet 5.5sonnet-5-5.md
Claude Haiku 5.5haiku-5-5.md
OpenAI GPT-6 Astragpt-6-astra.md
OpenAI GPT-6 Solgpt-6-sol.md
OpenAI GPT-6.1 Solgpt-6-1-sol.md
OpenAI GPT-6 Lunagpt-6-luna.md

Read the file matching the model actually running before applying this skill.

Running a model not listed above (an older generation, a different model family entirely)? sonnet-5.md carries the fullest general-purpose version of the underlying procedure — read that one rather than skipping this skill.

All versions