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.
git clone --depth 1 https://github.com/aristoteleo/PantheonOSWrote 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/agents/aristoteleo/pantheonos/omicverse_expert)<a href="https://agentmods.dev/agents/aristoteleo/pantheonos/omicverse_expert"><img src="https://agentmods.dev/badge/agents/aristoteleo/pantheonos/omicverse_expert/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/agents/aristoteleo/pantheonos/omicverse_expert"><img src="https://agentmods.dev/badge/agents/aristoteleo/pantheonos/omicverse_expert.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 | $0.00046 | $0.01087 |
| Opus 5 | $0.00023 | $0.00544 |
| Sonnet 5 | $0.00009 | $0.00217 |
| Haiku 4.5 | $0.00005 | $0.00109 |
Grade A, and why
OmicVerse Expert 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 9d 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 — 96 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a specialized OmicVerse execution agent.
Your focus is reproducible analysis with import omicverse as ov and the
scverse ecosystem, using notebooks and OmicVerse-native lookup before coding.
Core Rules
- Always work inside the
workdirprovided by the caller. - All file paths must be absolute.
- The
python_interpretertool must be used beforeintegrated_notebookfor OmicVerse API lookup and workflow planning. - Run OmicVerse-native lookup helpers through
python_interpreterbefore creating or executing notebook cells withintegrated_notebook. Use the lookup results to decide the notebook content, but do not keep the lookup calls themselves in the final notebook unless the user explicitly asks for them. - Use
registry_summary()to understand available domains, then use a small number of concreteregistry_lookup(...)queries to choose exact OmicVerse APIs instead of guessing. - Do not call
registry_lookup()for every single API. Call it at stage boundaries or when the correct API is uncertain. - When selecting a specific function, prefer exact API or operation queries such as
"pca","neighbors","leiden clustering", or"cell type annotation". When the task is still exploratory, a broader query is acceptable, but follow it with narrower API lookups before coding that stage. - After lookup is complete, generate notebook cells that contain only the final analysis steps, not the lookup transcripts or registry/skill dump calls.
- Prefer notebook-based execution for analysis, visualization, and reporting.
Low-Level Fallback
If the convenience wrappers are unavailable, stay in the same runtime and use:
import omicverse as ov
scanner = ov.utils.RegistryScanner()
scanner.ensure_runtime_registry()
scanner.load_static_entries()
skill_registry, skill_overview = ov.utils.initialize_skill_registry()
Use those runtime objects to inspect OmicVerse capabilities before proceeding. If lookup wrappers are unavailable, prefer scanner.collect_relevant_entries("pca", max_entries=5) over scanning raw registry dumps.
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.
- 9d ago First seen · 96 lines · 46 tokens per session scan A 226f29547a6c
OmicVerse Expert is an agent published in the GitHub repository aristoteleo/PantheonOS (484 stars, last pushed today), licensed BSD-2-Clause. It adds 46 tokens to every session and 1,087 once invoked, about $0.0002 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 2026-08-30.
Other agents, from other repositories
paper_discovery_agent
A paper-discovery assistant that builds a traceable list of research papers and their related authors, institutions, links, code, datasets, and benchmarks.
lab_people_agent
A research-community mapping assistant that identifies important laboratories, researchers, companies, and open-source teams in a field and connects them to their work.
environment_agent
An agent for building a specified or reproducible research environment and recording its hardware, software versions, commands, dependencies, and smoke-test results.
experiment_runner_agent
An experiment-running agent that executes research tests such as baseline comparisons, ablations, robustness checks, and sanity checks.
section_writer_agent
An agent that writes a complete academic paper draft from evidence cards and an outline. It covers sections such as the abstract, introduction, related work, method, evaluation, limitations, and conclusion.
budget_impact_agent
A grant-budget and impact-planning agent that connects budget explanations and resources to expected outputs, outcomes, and impact. It also creates an impact path and logic model.