with-relations

Validate typed skill relation predicates across hermetic v6 fixtures.

1|Updated May 6, 2026
One-click install
npx skills add https://github.com/jacob-balslev/skill-graph --skill with-relations
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: with-relations
Source: https://github.com/jacob-balslev/skill-graph/tree/main/examples/fixture-skills/with-relations
Command: npx skills add https://github.com/jacob-balslev/skill-graph --skill with-relations

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It verifies that the Skill Graph linting and relation-target resolver correctly handles all four typed relation predicates across a closed fixture cluster.

Core Features & Use Cases

  • Validates relation kinds together: Covers related, boundary, verify_with, and depends_on in a single v6 fixture.
  • Ensures hermetic cross-fixture resolution: Confirms relation targets resolve using only the fixtures directory (no dependency on a canonical sibling clone).
  • Enforces auditability: Requires non-empty reason fields so routing decisions remain traceable.

Quick Start

Run skill lint over the parent fixtures directory so relation targets can resolve across the sibling fixtures in the cluster.

Frequently Asked Questions about with-relations

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

FAQPage Schema
How do I validate typed relation predicates in a skill graph linting workflow?

Validating typed relation predicates requires resolving and verifying `related`, `boundary`, `verify_with`, and `depends_on` edges across a hermetic fixture set using the v6 `{skill, reason}` shape. This ensures deterministic resolver requirements are met.

Why does skill graph routing evaluation require non-empty reason fields in relation targets?

Routing evaluation requires non-empty reason fields to enforce auditability and ensure routing decisions remain traceable. This satisfies deterministic resolver requirements when relation targets resolve across sibling fixtures in the cluster.

What is the best way to lint cross-fixture dependencies without a canonical sibling clone?

Linting cross-fixture dependencies without a canonical sibling clone is achieved by ensuring relation targets resolve using only the fixtures directory. This hermetic resolution confirms cross-fixture targets are discoverable when scanning the fixture-suite root.

Can I test all four relation kinds in a single v6 fixture set?

Yes, you can validate `related`, `boundary`, `verify_with`, and `depends_on` relation kinds together in one v6 fixture set. This verifies the relation-target resolver correctly handles all typed predicates across a closed fixture cluster.

When do I need to use frontmatter schema v6 for relation predicate verification?

Frontmatter schema v6 is needed for relation predicate verification when applying linting and routing-eval scenarios that require the `{skill, reason}` shape to target sibling fixtures and ensure each relation includes a non-empty reason.