starwave-smolspec_skill

This skill creates a lightweight smolspec to rapidly plan small code changes, estimate scope, and outline tasks with clear success criteria.
  • Python

16

GitHub Stars

1

Bundled Files

3 weeks ago

Catalog Refreshed

2 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 veilstart where the catalogue uses aiagentskills.

npx veilstart add skill arjenschwarz/agentic-coding --skill starwave-smolspec

  • SKILL.md11.6 KB

Overview

This skill creates a compact, executable specification for small code changes that do not warrant the full spec workflow. It combines automated research, scope assessment, a proposed feature name, and lightweight documentation (smolspec and tasks) so engineers can implement and verify minor changes quickly. The skill enforces escalation rules and quality checks to avoid under-scoping complex work.

How this skill works

The skill first inspects the codebase to estimate complexity (LOC, file count, affected components) and looks for existing specs or related work. If the change fits micro-plan criteria, it generates a sanitized kebab-case feature name and waits for user approval. Then it produces two concise documents: specs/{feature_name}/smolspec.md and specs/{feature_name}/tasks.md, plus an optional decision_log.md when decisions are recorded.

When to use it

  • Fixing or adding a small feature expected under ~80 lines of code
  • Making a single-component change touching 1–3 files
  • Implementing a clear requirement that follows existing code patterns
  • Creating verifiable, developer-focused tasks for a minor change
  • When you need a minimal, self-contained spec for an AI or engineer to execute

Best practices

  • Run the skill only for isolated changes; escalate if multiple subsystems are affected
  • Provide a concise prompt and the preferred feature name when available
  • Approve or adjust the proposed feature name before proceeding
  • Review the generated smolspec.md for assumptions and clarify ambiguous requirements
  • Ensure tasks remain outcome-focused and include verification steps

Example use cases

  • Add a missing validation for a single form field following existing validation patterns
  • Introduce a toggle flag and implement small conditional logic in one module
  • Refactor a helper function used by a single component without wider API changes
  • Fix a rendering bug confined to a single view file
  • Add a small unit test and fixture for an isolated utility function

FAQ

If estimated LOC >80 or affected files >3, the skill recommends the full spec workflow and stops, explaining the findings and complexity metrics.

How is the feature name chosen?

The skill proposes a concise kebab-case name derived from your preference, the branch name, or the prompt, and waits for your approval before creating files.

Are the tasks implementation steps or outcomes?

Tasks are outcome-focused and verifiable. They describe what must be achieved, include success criteria, and distribute testing across tasks.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational