Audit linked-intent documents and codebase for inconsistencies and coverage.
npx skills add https://github.com/jmbeach/vat --skill arrow-maintenance-jmbeach
Please help me install this Agent Skill. Skill: arrow-maintenance Source: https://github.com/jmbeach/vat/tree/main/.claude/skills/arrow-maintenance Command: npx skills add https://github.com/jmbeach/vat --skill arrow-maintenance-jmbeach
The Arrow Maintenance Skill solves the complex challenge of managing and auditing linked-intent development, especially in projects too large to handle in a single context window.
Use the arrow-maintenance skill to start an audit and receive guidance by typing '/arrow-maintenance'.
High-intent search queries and answers about installing and using this skill.
You audit linked-intent documents by running a directed audit pass that checks existing LID documents and codebases for inconsistencies, coverage gaps, and coherence issues. This mode provides actionable updates to fix identified problems in large projects.
Linked-intent development connects project components across large codebases. You need an audit pass when managing these connections exceeds a single context window, requiring automated checks to maintain navigation and coherence across the project.
Yes, you can check linked-intent coherence during development using Python and Git. The skill operates via ambient mode to offer real-time suggestions and directed audit mode to perform thorough consistency checks on your existing LID documents.
The best way to keep linked-intent components in sync is using ambient guidance. This provides real-time suggestions for improvements while you work, ensuring all parts of your project stay aligned as changes occur.
Yes, linked-intent navigation works for projects too large for a single context window. The skill manages this complexity by operating on existing LID documents and the codebase directly to check for inconsistencies and coherence.
az9713Initialize a project with dark code prevention directories, CLAUDE.md section, and baseline audit.
imsaifGenerate targeted tests for UI components, hooks, and utilities to close coverage gaps.
ovachieverAutomate ML experiment tracking and model registry management via the wandb API.
testomatioAudit AI prompts and rules for contradictions, ambiguity, and incomplete guidance.
