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 skills add alex-alecu/garden-desk --skill legal-document-reviewgit clone --depth 1 https://github.com/alex-alecu/garden-deskWrote 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/alex-alecu/garden-desk/legal-document-review)<a href="https://agentmods.dev/skills/alex-alecu/garden-desk/legal-document-review"><img src="https://agentmods.dev/badge/skills/alex-alecu/garden-desk/legal-document-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/alex-alecu/garden-desk/legal-document-review"><img src="https://agentmods.dev/badge/skills/alex-alecu/garden-desk/legal-document-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 | $0.00000 | $0.00267 |
| Opus 5 | $0.00000 | $0.00133 |
| Sonnet 5 | $0.00000 | $0.00053 |
| Haiku 4.5 | $0.00000 | $0.00027 |
Grade A, and why
legal-document-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 12d 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.
What it actually says
Legal Document Review
Use supplied contract, parts, and playbook only. Do not assume jurisdiction, law, market standard, represented party, or missing value; ask party only if rank/negotiation changes. Check all parts for identity/signers, dates, economics, terms/duties, confidentiality/data/IP, liability/indemnity/insurance, assignment/change/termination/disputes, and references.
Internal inconsistencies first; one field/row: identity mismatch, value mismatch, date conflict, defined-term drift, clause conflict, broken reference, or missing repeated detail. A blank repeated signer name/title/authority/party/date is missing repeated detail. Report distinct values; aliases need source support.
Table: field, both sources/values, conflict, reason, and check. Cite both locations; absence cites expected place/checked parts. Never invent correction; separate fact, interpretation, missing information, and human decision; rank only with known party/goal. Do not edit unless requested. Qualified human review is required for legal decisions. This skill does not say a contract is valid, enforceable, compliant, or safe.
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.
- 12d ago First seen · 13 lines · 0 tokens per session scan A c212c7427dda
legal-document-review is a skill published in the GitHub repository alex-alecu/garden-desk (6 stars, last pushed yesterday), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 267 tokens. 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 skills, from other repositories
reporting-adverse-events
Structures adverse-event mentions that OpenMed extracts into FAERS / ICH E2B(R3) reportable fields — suspect drug, reaction (MedDRA PT), seriousness criteria, and outcome. Use when the user needs to build an individual case safety report (ICSR), populate a FAERS submission, map a narrative to E2B(R3) data elements…
auditing-deidentification-runs
Produce a signed, reproducible, no-PHI audit trail for an OpenMed de-identification run via deidentify(audit=True). Use when the user needs compliance evidence, a tamper-evident record of what was redacted and why, to verify nothing was changed, to retain proof for HIPAA/GDPR audits, or to review de-id decisions…
pseudonymizing-for-gdpr
Apply GDPR-grade pseudonymization to clinical or personal text with OpenMed, keeping a separately-held re-linkage key so the data can be controlled-re-linked later. Use when the user must process EU personal/health data under GDPR, asks for pseudonymization vs anonymization, needs Art. 4(5) / Art. 9 / Recital 26…
validating-us-core
Validate FHIR R4 resources and Bundles against US Core / USCDI profiles with the official HL7 FHIR validator before submitting to an EHR. Covers running validatorcli.jar (or the public validator.fhir.org), declaring meta.profile, must-support elements, common conformance gaps (missing code/category/status), and…
fda-database
Query openFDA API for drugs, devices, adverse events, recalls, regulatory submissions (510k, PMA), substance identification (UNII), for FDA regulatory data analysis and safety research.
setup-openmed
Collect a bounded set of de-identification policy decisions and write a deterministic, reviewable DEID-POLICY.md from the versioned local template. Use when a project needs explicit jurisdiction, recall floor, surrogate strategy, model policy, audit location, and human approval before privacy work begins.