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 agentmods add instructions/herry423/celltypepilot/agents-mdgit clone --depth 1 https://github.com/HERRY423/CellTypePilotWhat 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 | $0.04451 | $0.04451 |
| Opus 5 | $0.02226 | $0.02226 |
| Sonnet 5 | $0.00890 | $0.00890 |
| Haiku 4.5 | $0.00445 | $0.00445 |
Grade A, and why
CellTypePilot AGENTS.md 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 2d 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 — 319 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CellTypePilot — Codex Agent Instructions
Single-cell annotation plugin with evidence, conservative abstention, and reviewable draft output.
When to use
Activate CellTypePilot when the user wants to:
- Annotate, label, or identify cell types in pre-clustered scRNA-seq / spatial data
- Work with
.h5adfiles containing clustered single-cell data - Convert Seurat
.rdsfiles to.h5adfor analysis - Understand "what are these clusters?"
- Get auditable draft cell-type annotations with evidence for human review
- Review annotations interactively in a web panel
Trigger phrases: "annotate my clusters", "what cell types are these?", "label my scRNA-seq", "figure out the cell types", "run celltypepilot", "annotate my h5ad", "convert my rds", "review annotations in browser".
Prerequisites
Before running anything, verify the environment:
celltypepilot doctor
If core dependencies are missing, install them:
pip install celltypepilot
For a repository checkout used in development, pip install -e . remains appropriate.
CellTypePilot runs on CPU, requires Python >= 3.10, scanpy, anndata, and matplotlib. No MCP servers, no pixi, no conda environment needed for the basic path.
Optional extras:
pip install "celltypepilot[web]"— Web Inspector (Flask-based interactive review panel)pip install "celltypepilot[mcp]"— Native local CellTypePilot MCP facade for Agent hostspip install "celltypepilot[seurat]"— Seurat .rds conversion (requires rpy2 or R)pip install "celltypepilot[native]"— CellTypist, scANVI, popV (Python 3.12+), and bounded LLM Python runtimespip install "celltypepilot[all]"— All optional features
SingleR additionally requires R with the Bioconductor packages SingleR, zellkonverter,
scuttle, and SummarizedExperiment. CHETAH requires R with CHETAH, zellkonverter,
SummarizedExperiment, and SingleCellExperiment. CellAssign requires R with cellassign,
zellkonverter, and SummarizedExperiment, plus a frozen binary marker matrix.
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.
- 2d ago First seen · 319 lines · 4,451 tokens per session scan A e40148f9f289
CellTypePilot AGENTS.md is an instructions file published in the GitHub repository HERRY423/CellTypePilot (5 stars, last pushed 9d ago), licensed MIT. It adds 4,451 tokens to every session, about $0.0223 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 instructions, from other repositories
OmicsClaw AGENTS.md
Instructions for TianGzlab/OmicsClaw, covering agents.md — omicsclaw guide for ai coding agents, repository working contract, project overview, setup and pip install -e .
bioinformatics-agent-skills AGENTS.md
Instructions for YuliaNuzhnenko/bioinformatics-agent-skills, covering agent integration & specification guide 🤖, 📌 supported agent host architectures, 🧪 skill frontmatter specification and 🔒 security & sandboxing.
DrugClaw CLAUDE.md
Instructions for DrugClaw/DrugClaw, covering drugclaw, tech stack, directory overview, project layout and key patterns.
bio-gene-to-reference-tree copilot-instructions.md
Instructions for Hongda-Zhao/bio-gene-to-reference-tree, a project described as: Auditable agent skill for resolving protein queries, selecting references, and planning reproducible phylogenetic trees.
genefoundry-router CLAUDE.md
Instructions for berntpopp/genefoundry-router, a project described as: MCP gateway federating 21 biomedical MCP servers — gnomAD, ClinVar, HPO, UniProt, Ensembl VEP, PanelApp and more — behind one Streamable-HTTP endpoint, with collision-free namespaced tools and BM25 tool search.
uniprot-link CLAUDE.md
Claude Code instructions for berntpopp/uniprot-link, a project described as: MCP server for UniProt: protein function, sequence, features and cross-references over the UniProt SPARQL endpoint — intent-named tools plus a guarded raw-SPARQL escape hatch.