Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add bigbio/sdrf-skills/plugin install sdrf-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/bigbio/sdrf-skills/sdrf-adversarial-review)<a href="https://agentmods.dev/skills/bigbio/sdrf-skills/sdrf-adversarial-review"><img src="https://agentmods.dev/badge/skills/bigbio/sdrf-skills/sdrf-adversarial-review/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/bigbio/sdrf-skills/sdrf-adversarial-review"><img src="https://agentmods.dev/badge/skills/bigbio/sdrf-skills/sdrf-adversarial-review.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 | $0.00083 | $0.01514 |
| Opus 5.5 | $0.00033 | $0.00606 |
| Sonnet 5.5 | $0.00017 | $0.00303 |
| Haiku 4.5 | $0.00008 | $0.00151 |
Grade A, and why
sdrf-adversarial-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SDRF Adversarial Review
Bundle paths.
spec/,tools/anddata/ship with this skill, not with your working directory. Resolve every such path below against the bundle root (SDRF_SKILLS_ROOT, elseCLAUDE_PLUGIN_ROOT, else your sdrf-skills checkout):${SDRF_SKILLS_ROOT:-${CLAUDE_PLUGIN_ROOT}}/spec/sdrf-proteomics/TERMS.tsv. The helpers are thesdrf-toolscommand (install notes:sdrf-annotate/references/setup.md; check withsdrf-tools doctor); noPYTHONPATHor bundle-root variable is needed to run them. Files the user is annotating stay relative to the working directory.
Review the artifact from fresh context. Try to disprove its correctness; do not polish or extend it. Approve only what the artifact and cited evidence support.
The producer usually dispatches this review (sdrf-annotate Step 9.5). A caller
can also run it directly, for example as a separate pipeline task. The caller
then supplies the original request, the SDRF, the evidence manifest and the
deterministic validation output, and receives the findings and report.
Preserve Independence
- Confirm that this run is a fresh reviewer, hook-agent or separate-task context.
- Use only the original task, SDRF, specification, evidence manifest, repository files, and deterministic outputs supplied to the reviewer.
- Do not read or rely on the producer's transcript, reasoning, verdict, or summary. Treat producer claims as untrusted.
- If context isolation cannot be established, return
REVIEW_UNAVAILABLEand do not create a passing receipt. - Do not edit the SDRF. Return findings to the producer or caller that ran you.
- Read only files scoped to this artifact's accession. When reviewers run
concurrently, a shared scratchpad with generic filenames (
files_all.json,efetch.xml,mmc*.xlsx) silently substitutes one dataset's data for another's — the file still parses, it just describes a different PXD. Read only fromscratchpad/<PXD>/or the evidence paths the caller supplied, and assert on read: every fetched file list'sprojectAccessions, and every supplementary/efetchresult's returned title/accession, must match the artifact's accession before you use it — verify identity, not just HTTP 200.
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.
- yesterday Changed · +13 lines 0b5cfe41bd6a
- 1mo ago Changed 378e1bb64574
- 1mo ago Changed · +6 lines 9014e42d32a8
- 1mo ago First seen · 113 lines · 83 tokens per session scan A 06f15b871dfc
sdrf-adversarial-review is a skill published in the GitHub repository bigbio/sdrf-skills (19 stars, last pushed yesterday), licensed MIT. It adds 83 tokens to every session and 1,514 once invoked, about $0.0003 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 2026-08-30.
Other skills, from other repositories
proteomics-de
Load when computing two-group differential protein abundance (group2 vs group1, log2FC + p-value + BH-adjusted FDR) via Welch t-test, equal-variance t-test, or Mann-Whitney on a wide protein × sample CSV. Skip when you need multi-condition DE (run pairwise contrasts manually); label-based TMT linear-mixed models.
proteomics-enrichment
Load for Fisher over-representation analysis of protein identifiers against caller-supplied pathways. Skip rank-based GSEA (use bulkrna-enrichment) or protein differential testing (use proteomics-de).
proteomics-data-import
Load when ingesting a MaxQuant proteinGroups.txt, FragPipe combinedprotein.tsv, DIA-NN report, or generic CSV / TSV protein-quantification table — normalises columns to a standard schema, emits tables/proteins.csv. Skip when raw spectra are the input (run the search engine first); the file is already OmicsClaw schema.
proteomics-identification
Load when summarising peptide identifications (PSM count, unique peptide count, distinct protein count, score / charge distributions) from a peptide-level CSV produced by MaxQuant / FragPipe / DIA-NN. Skip when raw spectra are the input (run a search engine first); working with protein-quantification tables (use…
proteomics-ms-qc
Load when computing protein-table QC — proteins × samples count, missing-value rate, intensity CV (median + mean) — from a MaxQuant / FragPipe / DIA-NN protein-quantification CSV. Skip when raw mzML / RAW spectra are the input (run a search engine first); peptide-level QC is needed (use proteomics-identification).
proteomics-ptm
Load when summarising PTM sites (phosphorylation, acetylation, ubiquitination, etc.) from a per-site CSV — site-class assignment (Olsen et al. Class I/II/III by localizationprobability), per-PTM-type counts, amino-acid distribution, sites-per-protein. Skip when raw spectra are the input; you only need protein-level…