complete-code

Implements explicit production code stubs marked by TODO and TEMP sentinels within a given scope.

4|1|Updated Jul 31, 2025
One-click install
npx skills add https://github.com/alvis/.agents --skill complete-code-alvis
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: complete-code
Source: https://github.com/alvis/.agents/tree/main/plugins/coding/skills/complete-code
Command: npx skills add https://github.com/alvis/.agents --skill complete-code-alvis

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Codebases accumulate explicit implementation stubs—TODO(implementation) markers, TEMP placeholders, and draft-code sentinels—that must be finished without accidentally claiming neighboring work like bug fixes, test authoring, or new features. This Skill completes only those accepted stubs while routing everything else to the correct owning workflow. ## Core Features & Use Cases - Scoped stub completion: Scans a file, directory, or package for accepted markers and replaces each with the smallest production implementation satisfying the established contract. - Strict boundary routing: Classifies nearby markers and routes FIXME to coding:fix, test TODOs to coding:complete-test, and missing functionality to coding:write-code instead of overreaching. - Contract-driven verification: Reads specifications, types, call sites, and existing tests to derive behavior, then runs focused tests, type checks, and lint after each replacement. - Use Case: After running a draft-code pass that emitted throw new Error('IMPLEMENTATION: ...') sentinels across a package, invoke this Skill on that package to fill in each sentinel against the existing spec and test suite, with ambiguous markers reported as blocked rather than guessed. ## Quick Start Ask the agent to complete all implementation stubs in the src/payments directory using the complete-code workflow.

Frequently Asked Questions about complete-code

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I complete TODO implementation stubs in a codebase?

Provide a scope such as a file, directory, or package, and the workflow scans for accepted markers like TODO(implementation) and TEMP stubs. Each accepted stub is replaced with the smallest implementation satisfying the existing contract, verified by focused tests.

What markers does the complete-code workflow accept?

It accepts TODO(implementation) markers, legacy TODOs that unambiguously request production implementation, TEMP stubs with explicit replacement behavior, and the exact sentinel throw new Error('IMPLEMENTATION: ...') emitted by draft-code.

When should I use complete-code versus write-code?

Use complete-code only when an explicit stub exists and the surrounding contract already defines the behavior. Route missing functionality without a stub and newly requested features to write-code, which owns new functionality.

What happens when a stub's intended behavior is ambiguous?

Ambiguous markers are left untouched and reported as blocked rather than implemented by guessing. The same applies to stubs requiring uninstalled external dependencies or unspecified security-sensitive operations like authentication or cryptography.

Does complete-code write new tests for the implementations?

No, it only reads and runs existing tests to discover and verify the established contract. Test authoring, including it.todo and describe.todo markers, is routed to the complete-test workflow.