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/dualverse-ai/station/agents-mdgit clone --depth 1 https://github.com/dualverse-ai/stationWrote 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/dualverse-ai/station/agents-md)<a href="https://agentmods.dev/instructions/dualverse-ai/station/agents-md"><img src="https://agentmods.dev/badge/instructions/dualverse-ai/station/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 | $0.07598 | $0.07598 |
| Opus 5 | $0.03799 | $0.03799 |
| Sonnet 5 | $0.01520 | $0.01520 |
| Haiku 4.5 | $0.00760 | $0.00760 |
Grade B, and why
station AGENTS.md scanned grade B with 1 finding 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 5d 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
sudo apt install ripgrep How it starts
The opening of the file, as written. The whole thing — 536 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENT.md
This file gives Claude Code and other coding agents the current operating guidance for this repository.
First Principles
- Read the relevant docs before touching a subsystem. The
example/doc/folder contains subsystem notes. Use these as maintenance references, but verify against code when they disagree. - Check runtime overrides. Defaults live in
station/constants.py, butstation_data/constant_config.yamloverrides them at import time. Always inspect both when behavior depends on configuration. - Verify names and signatures before calling or editing. Use
rgto find definitions and call sites. Do not infer parameter names, constants, action names, or YAML fields from memory. - Treat
station_data/as live station state. Reading is allowed for debugging. Ask the user before modifying real station data unless the request explicitly asks for that modification. - Stay inside the current station checkout unless asked otherwise. When debugging live station state, only read or write the current repository and its own
station_data/orstation_multistart/paths. Do not inspect sibling checkouts such as~/station_2,~/station_3, or other~/station_*directories unless the user explicitly asks for a cross-station check or names that checkout. - Keep disposable local probes in
/tmp. Usetests/test_*.py,tests/debug_*.py, ortests/analysis_*.pyonly for files that should be visible to git. Use/tmpdirectly for generated API/debug snapshots, Sage/CAS probes, and other throwaway local test output. - Use repository file helpers for persistent station data. Use
station/file_io_utils.pyfor atomic YAML/text writes and safe directory creation.
Coding Agent Checklist
Before making code changes:
- Run
git status --shortand identify unrelated user changes. - Read this file, then read the subsystem doc listed in the Documentation Map.
- Inspect
station_data/constant_config.yamlwhen configuration affects the issue. - Use
rgto find the live implementation, constants, tests, and call sites. - Decide which files are code, docs, tests, or live station state before editing.
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.
- 5d ago First seen · 536 lines · 7,598 tokens per session scan B c8c294ab0fdb
station AGENTS.md is an instructions file published in the GitHub repository dualverse-ai/station (184 stars, last pushed 9d ago), licensed Apache-2.0. It adds 7,598 tokens to every session, about $0.0380 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
NanoResearch AGENTS.md
Instructions for OpenRaiser/NanoResearch, covering nanoresearch — codex integration mode, core goal, pipeline, workspace convention and paper modes.
NanoResearch CLAUDE.md
Instructions for OpenRaiser/NanoResearch, covering nanoresearch — claude code integration mode, core goal, pipeline, workspace convention and paper modes.
ai AGENTS.md
AGENTS.md instructions for vercel/ai, covering agents.md, project overview, repository structure, key directories and core package dependencies.
OwnPilot AGENTS.md
Instructions for ownpilot/OwnPilot, covering ownpilot, architecture, key patterns, commands and tech stack.
kernel-lore-mcp CLAUDE.md
Claude Code instructions for mjbommar/kernel-lore-mcp, covering kernel-lore-mcp — project state, pointers, standards, original goal and non-negotiable product constraints.
kent AGENTS.md
AGENTS.md instructions for respawn-llc/kent, covering repository layout, product principles, tooling, frozen rust code and coding guidelines & memories.