impl_skill

This skill helps you implement features and write code based on Plans.md tasks, ensuring quality gates and non hard-coded solutions.
  • Shell

212

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill chachamaru127/claude-code-harness --skill impl

  • SKILL.md9.5 KB

Overview

This skill implements features and writes JavaScript code based on tasks registered in Plans.md. It enforces strict quality guardrails to avoid hardcoding, stubs, or test-driven shortcuts, and integrates LSP guidance and optional memory-based pattern lookup to produce robust, maintainable implementations.

How this skill works

Before starting, the skill confirms the task exists in Plans.md and adds it if missing. It runs a quality gate to decide TDD, security, and performance considerations, optionally searches past implementation patterns when memory is enabled, and then implements code following referenced implementation and testing guidelines while using LSP for context and verification.

When to use it

  • When the user asks to implement a feature or add a new function in JavaScript
  • When the user requests writing production-ready code (not reviews or build verification)
  • When a task is recorded or needs to be recorded in Plans.md before coding
  • When tests should be created or updated alongside the implementation
  • When LSP tools are available to inspect related code and types

Best practices

  • Always confirm or add the task in Plans.md before implementing
  • Run the quality gate: decide if TDD, security checks, or performance concerns apply
  • Never use hardcoded answers, stub returns, or test-only mappings
  • Use LSP (go-to-definition, find-references, hover, diagnostics) to understand and validate changes
  • Handle edge cases (null/empty/boundaries) and make logic general-purpose
  • If implementation is unclear, report difficulties with concrete options instead of committing a placeholder

Example use cases

  • Add a new service function under src/services/ and create corresponding unit tests
  • Implement a core business logic feature in src/core/ with TDD suggested
  • Create input validation utilities for API routes and flag security checks
  • Follow past patterns (if memory enabled) when implementing a similar API endpoint
  • Use LSP to update a function implementation after locating its type definitions and references

FAQ

Report the blocking points honestly with a short consultation note listing the situation, specific difficulties, and possible options instead of writing an incomplete stub.

Do you write tests as well?

Yes — the quality gate will recommend TDD where appropriate, and tests should be created or updated as part of the implementation workflow.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational
impl skill by chachamaru127/claude-code-harness | VeilStrat