Borrowing it
Nothing to install: this file belongs to eclipse-rdf4j/rdf4j. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/eclipse-rdf4j/rdf4j/main/.agent/skills/jmh-benchmark-compare/SKILL.mdgit clone --depth 1 https://github.com/eclipse-rdf4j/rdf4jWrote 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/eclipse-rdf4j/rdf4j/jmh-benchmark-compare)<a href="https://agentmods.dev/skills/eclipse-rdf4j/rdf4j/jmh-benchmark-compare"><img src="https://agentmods.dev/badge/skills/eclipse-rdf4j/rdf4j/jmh-benchmark-compare/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/eclipse-rdf4j/rdf4j/jmh-benchmark-compare"><img src="https://agentmods.dev/badge/skills/eclipse-rdf4j/rdf4j/jmh-benchmark-compare.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 | $0.00100 | $0.00792 |
| Opus 5.5 | $0.00040 | $0.00317 |
| Sonnet 5.5 | $0.00020 | $0.00158 |
| Haiku 4.5 | $0.00010 | $0.00079 |
Grade A, and why
jmh-benchmark-compare 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 1mo 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 — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
jmh-benchmark-compare
Use this skill when benchmark output comparison must be reproducible, sortable, and exportable.
Quick start
Run two-file comparison:
python3 .codex/skills/jmh-benchmark-compare/scripts/jmh_benchmark_compare.py \
/path/run-a.txt /path/run-b.txt \
--export-formats txt,md,csv,xlsx,html \
--output-dir /tmp \
--output-base jmh-compare
Sort by diff percent (descending):
python3 .codex/skills/jmh-benchmark-compare/scripts/jmh_benchmark_compare.py \
run-a.txt run-b.txt \
--sort-column "Diff % [run-b - run-a]" \
--sort-desc \
--export-formats md \
--output /tmp/jmh-diff.md
Core behavior
- Detect first JMH table header line:
line.startswith("Benchmark") and "Mode" in line and "Score" in line. - Derive column boundaries from that header.
- Parse all following benchmark rows into an internal table.
- Match overlapping benchmark keys across files.
- Add derived columns:
Diff Score [target - baseline],Diff % [target - baseline],Status [...].
Default key columns are all columns except Cnt, Score, Error. Override via --id-columns.
Inputs and overlap
- Pass any mix of files and directories.
- Directory entries are scanned for files that contain a JMH header.
--overlap-mode allkeeps only rows present in all files.--overlap-mode anykeeps rows present in at least two files.- Baseline selection:
--baseline <index-or-label>.
Filters and regression shortcuts
- Hide tiny deltas:
--min-deviation-pct 1.0 - Show only regressions above threshold:
--regressions-over-pct 3.0 - Control direction interpretation:
--score-direction auto|higher|lower
Historical analysis
Analyze trends across many runs:
python3 .codex/skills/jmh-benchmark-compare/scripts/jmh_benchmark_compare.py \
/path/bench-history \
--recursive \
--glob "*.txt" \
--timestamp-source auto \
--analyze-over-time \
--regressions-over-pct 2.5 \
--export-formats html,csv \
--output-dir /tmp \
--output-base jmh-history
What ships with it
6 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.
- 1mo ago First seen · 98 lines · 100 tokens per session scan A ef9fc56ee748
jmh-benchmark-compare is a skill published in the GitHub repository eclipse-rdf4j/rdf4j (422 stars, last pushed today), licensed BSD-3-Clause. It adds 100 tokens to every session and 792 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-09-03.
Other skills, from other repositories
open-ontologies
AI-native ontology engineering using 110 MCP tools backed by an in-memory Oxigraph triple store. Build, validate, query, and govern RDF/OWL ontologies with a generate-validate-iterate loop. Use when building ontologies, knowledge graphs, RDF data, SPARQL queries, BORO/4D modeling, SHACL validation, clinical…
mcp-sparql
Exposes SPARQL query capabilities to LLMs via the Model Context Protocol. Supports SELECT, ASK, CONSTRUCT, and DESCRIBE queries against any SPARQL endpoint.
knowledge-graph-sparql
Knowledge graph construction, SPARQL querying, and entity linking for library and research data management using RDF and Wikidata.
prism
PRISM finds entities (genes, proteins, compounds) at the INTERSECTION of two or more property sets on RDF knowledge graphs via TogoMCP — e.g. "genes common to disease X and druggable pathway Y", "targets both linked to phenotype P and modulated by existing drugs". Use whenever the user asks a set-intersection /…
research-article-analysis
Systematically validate a research article's claims against TogoMCP RDF databases — molecular formulas, reaction equations, pathways, protein function, GO definitions — instead of trusting the paper's text or keyword-search snippets. Use whenever the user asks to "analyze this paper/article", fact-check or validate a…
disease-analysis
Run a multi-scale disease pathophysiology analysis — molecular defects through to clinical symptoms — using TogoMCP (RDF Portal SPARQL), TogoID ID conversion, OLS4, and PubMed. Use whenever the user asks to "analyze a disease", build a disease mechanism model, find disease-associated proteins/pathways/drugs, map a…