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 commands/mims-harvard/tooluniverse/comparegit clone --depth 1 https://github.com/mims-harvard/ToolUniverseWrote 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/commands/mims-harvard/tooluniverse/compare)<a href="https://agentmods.dev/commands/mims-harvard/tooluniverse/compare"><img src="https://agentmods.dev/badge/commands/mims-harvard/tooluniverse/compare.svg" alt="Measured on agentmods" 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 | $0.00074 | $0.01365 |
| Opus 5 | $0.00037 | $0.00682 |
| Sonnet 5 | $0.00015 | $0.00273 |
| Haiku 4.5 | $0.00007 | $0.00136 |
Grade A, and why
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 4d 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 — 119 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Compare these items side-by-side: $ARGUMENTS
Comparison ≠ profile-each-then-merge. The structure has to be decided up front (what columns? what dimensions?), the data has to be aligned (same units, same date range, same reference), and the output has to be a TABLE the user can act on, not a stack of paragraphs.
Process
1. Detect domain
Inspect the items to figure out what KIND of things they are:
| Items look like | Domain | Column set |
|---|---|---|
| Generic drug names, kinase inhibitors, mAbs | drug | mechanism, target(s), approved indications, key trials, AE profile, FDA status, dosing |
| Gene symbols (BRCA1, TP53, KRAS) | gene/target | function, disease associations, druggability, # approved drugs targeting, mutation frequency in cancer |
| Disease names, syndromes | disease | prevalence, key genes, treatments, ICD codes, clinical trials count |
| Variants (V600E, R175H) | variant | gene, amino acid change, pathogenicity, frequency (gnomAD), drug sensitivity, evidence tier |
| Clinical trial NCT IDs | trial | phase, indication, intervention, enrollment, status, primary endpoint, sponsor |
| Cell lines | cell line | tissue, mutations, drug sensitivity, source |
If items mix domains (e.g., "BRCA1, alpelisib"), ask the user what comparison axis they want — drug-vs-target makes no sense as a row-aligned table.
If domain isn't obvious, look up the first item: find_tools with the item
name, see what kind of tool matches. Drug-tools → drug; gene-tools → gene; etc.
State the detected domain in one line before continuing.
2. Decide the column set
Use the table above as the default. Ask yourself: what's the user actually trying to PICK BETWEEN? Drop columns that don't help discriminate (e.g., if all items have the same FDA status, drop that column). Keep 5-8 columns max — more becomes unreadable.
If the user's prompt gave a hint ("compare on safety"), bias the column set toward that axis (more safety columns, fewer mechanism columns).
3. Gather data per item, per column
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.
- 4d ago First seen · 119 lines · 74 tokens per session scan A 41e2a1303349
compare is a command published in the GitHub repository mims-harvard/ToolUniverse (1,670 stars, last pushed 2d ago), licensed Apache-2.0. It adds 74 tokens to every session and 1,365 once invoked, about $0.0004 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 commands, from other repositories
git
Git operations with intelligent commit messages and workflow optimization.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.