Incremental Implementation
Makes the agent build features in small, tested, committable slices instead of one big pass.
- Repository
- addyosmani/agent-skills
- Stars
- 112,668
- Licence
- MIT
- Files
- 1 file
What it does
When installed, the agent works through larger coding tasks in small steps: implement one thin piece, run tests and the build, confirm it works, commit, then move on. It also follows rules like keeping changes minimal, avoiding unrequested cleanup, and using feature flags for unfinished work. The skill is a set of written instructions for the agent; it does not add new tools on its own.
Great for
- Developers who want an AI agent to make multi-file changes in reviewable, working increments
- Teams following a task plan who want each task landed and verified step by step
- Anyone whose agent tends to write a large amount of code before running any tests
Skip it if
- Single-file or single-function edits, where the skill itself says it is unnecessary overhead
- Projects with no test suite or build, since the verification steps assume existing commands to run
- Users who expect the skill to write code in one large pass quickly; it intentionally slows work down into slices
Try asking
- “Implement Task 3 from our plan, starting with the database schema and API only, and run the tests after.”
- “Add user sharing to this feature behind a feature flag, landing it in small commits.”
- “This refactor touches many files; break it into slices and verify each one before continuing.”
- “Build the task list feature in vertical slices: create, list, edit, then delete.”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 1 original files, the MIT licence and an attribution note.
incremental-implementation.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip incremental-implementation.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip incremental-implementation.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Give the agent a clear plan or task list first, and state explicitly what is in and out of scope for each increment.
- The skill tells the agent to run your repository's own test, build, and lint commands, so make sure those exist and work.
- Committing after each slice relies on the separate git-workflow-and-versioning skill for good commit guidance; install it too if you want that.
What's inside
- SKILL.md
Redistributed unmodified from addyosmani/agent-skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.