What is SetOptics AGENTS.md?

Repository-specific instructions for SetOptics, an open-source project that detects volleyball rallies in video. They explain the project layout, approved entry points, and working rules for coding agents.

Working on SetOptics detection pipelines, annotation tools, scripts, and rally or segment evaluation tasks.

They give an agent the local context needed to change or run the project without guessing where code belongs or which evaluation tools to use.

SetOptics AGENTS.md — Instructions file by dawsonpar

Install

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.

agentmods

npx agentmods add instructions/dawsonpar/setoptics/agents-md

Clone the repo

git clone --depth 1 https://github.com/dawsonpar/SetOptics

Made for: Codex, OpenCode.

Per session 1,173 This file is loaded in full into every session.

When invoked 1,173 The same file — it is already loaded in full.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

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	$0.01173	$0.01173
Opus 5	$0.00587	$0.00587
Sonnet 5	$0.00235	$0.00235
Haiku 4.5	$0.00117	$0.00117

Measured 2d ago against content hash 1b38d94c02b5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

SetOptics 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.

AGENTS.md · 118 lines

How it starts

The opening of the file, as written. The whole thing — 118 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

Conventions for any agent (Claude Code, Cursor, Codex, Aider, autonomous or interactive) working inside this repository. Harness-agnostic.

What this repo is

Open-source rally-detection pipelines for volleyball video. Library + scripts, no service layer. The detection modes are: signal, videomae, ensemble, llm (Gemini).

Where the code lives

setoptics/ is the importable Python package (ball tracker, Gemini pipeline, prompts, base classes).
scripts/ holds standalone CLI entry points. They are research-grade but functional. Each one is its own __main__ and uses argparse.
tools/annotation/ holds the ground-truth labeling pipeline. It is importable and runnable.
tools/shared/eval/temporal.py is the canonical evaluation framework. Use it for any new rally/segment metric work; do not write parallel metric code.

Working rules for agents

Read README.md first. It is the source of truth for what each mode does and how to invoke it.
Use the canonical entry points. Detection runs through scripts/*_rally_detector.py. Annotation runs through tools/annotation/annotate_sliding_window.py (high quality) or annotate_fast.py (fast iteration). Do not invent new wrappers.
Activate the venv before running Python. All scripts assume the project venv is active (source .venv/bin/activate).
Do not commit user data. data/ and any *_annotations*.json files are gitignored. Do not stage them.
One file = one purpose. Hard cap of 500 lines per code file. If a change pushes over this, split.
No new directories without reason. The layout is fixed; new modules go in setoptics/, new CLIs in scripts/, new annotation tools in tools/annotation/.
Apache 2.0. All new files inherit the repo license. Do not add conflicting license headers.

Common tasks

Run rally detection on a video

python scripts/signal_rally_detector.py --video FOOTAGE.mp4 --output OUT.json
python scripts/ensemble_rally_detector.py --video FOOTAGE.mp4 --output OUT.json --mode fast
python tools/annotation/annotate_sliding_window.py FOOTAGE.mp4   # LLM path

Read the full file on GitHub · 118 lines

Changes

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 · 118 lines · 1,173 tokens per session scan A 1b38d94c02b5

Subscribe to this mod's changes

SetOptics AGENTS.md is an instructions file published in the GitHub repository dawsonpar/SetOptics (24 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 1,173 tokens to every session, about $0.0059 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.

spec-kit AGENTS.md

AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.

github/spec-kit · 7,104 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

SetOptics AGENTS.md