Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add AnthonyAlcaraz/agentic-graph-rag-skills --skill graph-model-selectorgit clone --depth 1 https://github.com/AnthonyAlcaraz/agentic-graph-rag-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/anthonyalcaraz/agentic-graph-rag-skills/graph-model-selector)<a href="https://agentmods.dev/skills/anthonyalcaraz/agentic-graph-rag-skills/graph-model-selector"><img src="https://agentmods.dev/badge/skills/anthonyalcaraz/agentic-graph-rag-skills/graph-model-selector/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/anthonyalcaraz/agentic-graph-rag-skills/graph-model-selector"><img src="https://agentmods.dev/badge/skills/anthonyalcaraz/agentic-graph-rag-skills/graph-model-selector.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00225 | $0.02176 |
| Opus 5 | $0.00112 | $0.01088 |
| Sonnet 5 | $0.00045 | $0.00435 |
| Haiku 4.5 | $0.00022 | $0.00218 |
Grade A, and why
graph-model-selector scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 144 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Graph Model Selector
Overview
AI agents are only as intelligent as the knowledge structures you give them, and the model you choose constrains what they can reason about. The chapter is explicit: start with your reasoning requirements, not your data. Formal logical inference (medical diagnosis) demands a different model than traversal-heavy network analysis or many-entity events.
Three model classes, each with a characteristic profile:
- Labeled property graph (LPG): fast traversals, mature ecosystem (Neo4j / Neptune / ArangoDB), flexible. Can only tell you what you explicitly programmed — no native inference.
- RDF (subject-predicate-object triples): formal logical semantics enable
native inference. Tell it
Disease1 causes Symptom1andPatient exhibits Symptom1and it infersDisease1is a candidate diagnosis with no rule written. Slower; n-ary needs reification. - Hypergraph: one hyperedge connects any number of entities, so n-ary relations are native — no auxiliary nodes. Ecosystem is immature.
The selector scores each model across the five features the chapter names (formal reasoning, n-ary, performance, tool ecosystem, constraint expressiveness) weighted by the caller's requirements, and surfaces a hybrid recommendation when the top two are close (the chapter's "Putting it all together" hybrid guidance).
The n-ary helper makes the trade-off concrete. A prescription
(doctor + patient + medication + dosage + date + condition) is one hyperedge
(representation_cost.hypergraph_elements == 1) but costs 1 + arity elements
(one relation node + one edge per participant) under LPG or RDF reification.
When to Use
- Choosing a graph backend model class for a new agentic system
- Justifying a build-vs-buy or LPG-vs-RDF-vs-hypergraph decision in a design doc
- Deciding whether a multi-entity fact warrants a hyperedge or reification
Phrases: "which graph model", "property graph vs RDF", "hypergraph", "n-ary relation", "reification", "graph data model selection", "Neo4j vs RDF".
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 11d ago First seen · 144 lines · 225 tokens per session scan A f6f4a4e8156c
graph-model-selector is a skill published in the GitHub repository AnthonyAlcaraz/agentic-graph-rag-skills (10 stars, last pushed 2mo ago), licensed MIT. It adds 225 tokens to every session and 2,176 once invoked, about $0.0011 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
embeddings
Vector embeddings with HNSW indexing, sql.js persistence, and hyperbolic support. 75x faster with agentic-flow integration. Use when: semantic search, pattern matching, similarity queries, knowledge retrieval. Skip when: exact text matching, simple lookups, no semantic understanding needed.
postgresql-table-design
Use this skill when designing or reviewing a PostgreSQL-specific schema. Covers best-practices, data types, indexing, constraints, performance patterns, and advanced features.
cqrs-implementation
Implement Command Query Responsibility Segregation for scalable architectures. Use when separating read and write models, optimizing query performance, or building event-sourced systems.
projection-patterns
Build read models and projections from event streams. Use when implementing CQRS read sides, building materialized views, or optimizing query performance in event-sourced systems.
stripe-projects
Use when the user wants to provision infrastructure or third-party services using Stripe Projects. Triggers: "I need a database", "set up auth", "add caching", "give me a Postgres", "provision Redis", "I need hosting", "add a vector DB", "get me an API key for X", "get credentials for X", "sign up for a service", "set…
prisma-postgres-setup
Set up a new Prisma Postgres database and connect it to a local project using the Management API. Use when asked to "set up a database", "create a Prisma Postgres project", "get a connection string", "connect my app to Prisma Postgres", or "provision a database".