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/cqframework/clinical-reasoning/agents-mdgit clone --depth 1 https://github.com/cqframework/clinical-reasoningWhat 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 | $0.00927 | $0.00927 |
| Opus 5 | $0.00464 | $0.00464 |
| Sonnet 5 | $0.00185 | $0.00185 |
| Haiku 4.5 | $0.00093 | $0.00093 |
Grade A, and why
clinical-reasoning AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Project Overview
This is the CQF Clinical Reasoning on FHIR repository, which provides JVM-based, FHIR-enabled clinical reasoning components. The project implements FHIR Clinical Reasoning operations like Measure evaluation ($evaluate-measure), PlanDefinition application ($apply), and CQL evaluation capabilities.
See the ARCHITECTURE.md document for a detailed overview of the vision, architecture, design patterns, and coding standards used in this project.
Build and Test Commands
Building the Project
# Full build with tests
./gradlew build
# Build without tests
./gradlew assemble
# Compile only
./gradlew compileJava
# Parallel build (enabled by default via gradle.properties)
./gradlew build
Running Tests
# Run all tests
./gradlew test
# Run tests in a specific module
./gradlew :cqf-fhir-cr:test
# Run a single test class
./gradlew :MODULE:test --tests ClassName
# Run a single test method
./gradlew :MODULE:test --tests "ClassName.methodName"
# Run integration tests
./gradlew integrationTest
# Skip tests
./gradlew assemble
Code Formatting and Quality
# Check code formatting (runs in CI on PRs)
./gradlew spotlessCheck
# Apply code formatting fixes
./gradlew spotlessApply
# This project uses Palantir Java Format (configured in buildSrc convention plugins)
# Checkstyle runs automatically as part of the build
Linked Builds (Cross-Repo Development)
To build against a local checkout of a dependency (e.g., CQL Engine) instead of Maven Central:
# Copy the template and set paths to your local checkouts
cp local.properties.example local.properties
# Edit local.properties — uncomment and set the path:
# cql.engine.path=../clinical_quality_language/Src/java
# Build as normal — the settings plugin detects local.properties and includes the linked build
./gradlew build
Available linked builds are listed in local.properties.example. The registry is defined in build-logic/src/main/kotlin/cqf/LinkedBuild.kt.
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 · 129 lines · 927 tokens per session scan A ef28b8f2e8a9
clinical-reasoning AGENTS.md is an instructions file published in the GitHub repository cqframework/clinical-reasoning (49 stars, last pushed 4d ago), licensed Apache-2.0. It adds 927 tokens to every session, about $0.0046 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 instructions, from other repositories
Sighthound AGENTS.md
Instructions for Corgea/Sighthound, covering claude, commands, python / django test naming and agent skill.
wisp-science AGENTS.md
AGENTS.md instructions for xuzhougeng/wisp-science, covering agents.md, project orientation, repository layout, engineering rules and verification commands.
MoleCode AGENTS.md
Instructions for AtomFlow-AI/MoleCode, covering agents.md — molecode, what this repo is, use the molecode skill, molecode-first rule and cli quick start.
EDDI AGENTS.md
Instructions for labsai/EDDI, covering eddi backend — ai agent instructions, 1. project context, ecosystem, key architecture and build & test commands.
tachyon GEMINI.md
Instructions for kpavlov/tachyon: In repositories indexed by CodeGraph (a .codegraph/ directory exists at the repo root), reach for it BEFORE grep/find or reading files when you need to understand or locate code.
relarena CLAUDE.md
Instructions for PriorLabs/relarena: All package notes live in AGENTS.md — read that first.