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/fl-sean03/opensdl/agents-mdgit clone --depth 1 https://github.com/fl-sean03/OpenSDLWrote 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/instructions/fl-sean03/opensdl/agents-md)<a href="https://agentmods.dev/instructions/fl-sean03/opensdl/agents-md"><img src="https://agentmods.dev/badge/instructions/fl-sean03/opensdl/agents-md.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 | $0.00916 | $0.00916 |
| Opus 5 | $0.00458 | $0.00458 |
| Sonnet 5 | $0.00183 | $0.00183 |
| Haiku 4.5 | $0.00092 | $0.00092 |
Grade A, and why
OpenSDL 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 3d 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 — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repository instructions
Project
OpenSDL is a modular framework for computational and autonomous laboratories.
Layout
- reusable packages:
packages/ - deployable applications:
apps/ - integrations:
adapters/ - scientific extensions:
domain-packs/ - complete examples:
examples/ - cross-package tests:
tests/
Commands
- install:
uv sync --locked --all-packages --group dev - test:
make test— workspace suite and the digital-twin surrogate overlay - lint:
make lint— lockfile, Ruff lint and format, Pyright, boundaries, schema drift, repository, and versions - viewer:
make viewer— surrogate viewer lint, typecheck, tests, build, and committed asset drift - docs:
make docs—mkdocs build --strict - example:
make example— the complete simulated campaign - format:
make format— applies Ruff formatting and safe fixes;make lintenforces it - schemas:
uv run --locked python scripts/generate-schemas.py
make test, make lint, make viewer, make docs, make example, and make showcase together
cover every check the pull-request CI job enforces. make showcase re-derives the campaign the
README frame is rendered from, which takes about a minute and is why CI runs it as its own job.
make scene covers the one that runs separately: the headless
Blender rebuild proving the committed scene is reproducible from source. It needs the exact Blender
version recorded in the scene's node inventory and takes several minutes. A bare
uv run --locked pytest is not the full suite: testpaths excludes examples/, so the surrogate
tests are reachable only through make test or make surrogate, and the showcase campaign only
through make showcase. Narrower targets (unit, integration, e2e, conformance,
typecheck, boundaries, validate, showcase, surrogate, propagation, scene) and the raw
command behind each target are in the Makefile.
Architecture rules
coreimports no internal package.- Applications compose packages; business logic stays in packages.
- Vendor or facility behavior belongs in adapters.
- Every operational adapter needs simulation and conformance coverage.
- Public models are typed and exported as versioned schemas.
- Database access goes through repository interfaces.
- A change is complete when code, tests, schemas, examples, and documentation agree.
- A laboratory with one instrument stays expressible in about fifteen manifest lines, runnable in
one process against SQLite, with no scheduler, broker, or optional service. Facility features are
opt-in by configuration, never by requirement, and a change that lengthens the minimum manifest
needs justifying.
tests/test_minimal_laboratory.pyenforces this.
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.
- 3d ago First seen · 74 lines · 916 tokens per session scan A eaeb8a97144f
OpenSDL AGENTS.md is an instructions file published in the GitHub repository fl-sean03/OpenSDL (102 stars, last pushed 3d ago), licensed Apache-2.0. It adds 916 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
scitex-python CLAUDE.md
Claude Code instructions for scitex-ai/scitex-python, covering workflow orchestration, 1. plan mode default, 2. subagent strategy, 3. self-improvement loop and 4. verification before done.
scholaraio AGENTS.md
Instructions for ZimoLiao/scholaraio, covering scholaraio - agent entry, what scholaraio is, how to work in this repo, start here and skill-first workflow.
scholaraio CLAUDE.md
Instructions for ZimoLiao/scholaraio: This file is the Claude Code project-memory entrypoint. It intentionally stays light.
scholaraio copilot-instructions.md
Instructions for ZimoLiao/scholaraio, covering scholaraio — github copilot instructions and quick reference.
PyAutoArray AGENTS.md
Instructions for PyAutoLabs/PyAutoArray, covering pyautoarray — agent instructions, what this repo is, related repos, architecture and quick commands.
mcs CLAUDE.md
Instructions for mcs-cli/mcs, covering claude.md, project overview, commands, development and cli usage (after install).