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 Knuckles-Team/epistemic-graph --skill kg-modality-reasoninggit clone --depth 1 https://github.com/Knuckles-Team/epistemic-graphWrote 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/knuckles-team/epistemic-graph/kg-modality-reasoning)<a href="https://agentmods.dev/skills/knuckles-team/epistemic-graph/kg-modality-reasoning"><img src="https://agentmods.dev/badge/skills/knuckles-team/epistemic-graph/kg-modality-reasoning/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/knuckles-team/epistemic-graph/kg-modality-reasoning"><img src="https://agentmods.dev/badge/skills/knuckles-team/epistemic-graph/kg-modality-reasoning.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00090 | $0.00690 |
| Opus 5 | $0.00045 | $0.00345 |
| Sonnet 5 | $0.00018 | $0.00138 |
| Haiku 4.5 | $0.00009 | $0.00069 |
Grade A, and why
kg-modality-reasoning 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 10d 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.
What it actually says
kg-modality-reasoning — OWL / rule inference in the engine
The engine reasons natively over its own RDF/OWL store (eg-rdf/owl). It performs OWL 2 EL⁺
completion (sub/conj/some/chain/subrole/bot/disjoint), OWL 2 RL property rules
(transitive/symmetric/inverse/chains/domain), classification and consistency (an unsatisfiable
class ⇒ inconsistent), and forward-chaining materialization with incremental add_axioms
(a monotone fixpoint resumed in place). DL-requiring ontologies route to a pure-Rust
description-logic tableau (owl-dl: cardinality, complementOf, nominals) while the
EL/RL fast path stays default. SWRL user rules add a Horn-rule DSL + swrlb: built-ins.
Confidence-weighting (per-axiom eg:confidence, noisy-OR), Ebbinghaus time-decay, and
distributed/cross-shard reasoning (one closure over a unioned TBox+ABox) are supported.
Reasoning is also a query-time op (Op::Reason under owl-plan seeds a RowSet). See
docs/capabilities.md → OWL reasoning (eg-rdf/owl).
The MCP way (through graph-os)
load_tools(tools=["engine_reasoning"]) # then classify / materialize / check-consistency
or the REST twin graph-os exposes for the reasoning modality.
The wire way
Reasoning composes with the SPARQL surface (kg-modality-sparql): ICV integrity-constraint
validation can run over the OWL-reasoned view, and a reasoned RowSet is reachable from a
plan via Op::Reason. There is no separate reasoning port — it operates on the triples served
by /sparql (EPISTEMIC_GRAPH_SPARQL_ADDR).
Cross-modal seam
Materialized inferences are written back into the same store the SQL/RDF/vector modalities
read, so a reasoned closure is visible to kg-modality-sql and kg-modality-sparql and is
committed under the engine's unified cross-modal transaction — inference is a first-class
citizen of the shared substrate, not a bolt-on batch job.
Related
kg-modality-sparql— the RDF store the reasoner classifies and materializes over.
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.
- 10d ago First seen · 53 lines · 90 tokens per session scan A 0462114f7789
kg-modality-reasoning is a skill published in the GitHub repository Knuckles-Team/epistemic-graph (10 stars, last pushed 12d ago), licensed MIT. It adds 90 tokens to every session and 690 once invoked, about $0.0005 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
data-engineer
Build scalable data pipelines, modern data warehouses, and real-time streaming architectures. Implements Apache Spark, dbt, Airflow, and cloud-native data platforms.
groq-inference
Ultra-fast LLM inference on custom LPU hardware. OpenAI-compatible API at api.groq.com. Lowest latency in the industry (500-1000+ tok/s). Supports chat completions, vision, audio (Whisper STT + TTS), tool calling, JSON mode, and streaming. Free tier available. Inference only — no training.
graphjin-env
Use when setting up a training or evaluation loop against a GraphJin agent environment — running the container, reading /health, driving episodes hosted or step-by-step or with your own agent over MCP, splitting train from eval, exporting trajectories, and deciding whether two rewards can be compared.
similarity-search-patterns
Implement efficient similarity search with vector databases. Use when building semantic search, implementing nearest neighbor queries, or optimizing retrieval performance.
ingesting-into-data-lake
Import data into the AWS data lake from S3 files, local uploads, JDBC databases (Oracle, SQL Server, PostgreSQL, MySQL, RDS, Aurora), Amazon Redshift, Snowflake, BigQuery, DynamoDB, or existing Glue catalog tables (migration). Default target is S3 Tables; standard Iceberg on a general purpose bucket is supported where…
nornicdb-qdrant-migration
Migrate from Qdrant to NornicDB end-to-end through NornicDB's Qdrant-compatible gRPC surface. Covers connection setup, collection→database mapping, point→node mapping, the vector-config and named-vector replication, point upsert in batches, count verification, and what (deliberately) does not transfer (snapshots, HNSW…