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/moeru-ai/auv/agents-mdgit clone --depth 1 https://github.com/moeru-ai/auvWrote 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/moeru-ai/auv/agents-md)<a href="https://agentmods.dev/instructions/moeru-ai/auv/agents-md"><img src="https://agentmods.dev/badge/instructions/moeru-ai/auv/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 | $0.05903 | $0.05903 |
| Opus 5 | $0.02951 | $0.02951 |
| Sonnet 5 | $0.01181 | $0.01181 |
| Haiku 4.5 | $0.00590 | $0.00590 |
Grade A, and why
auv 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 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.
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 — 536 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AUV Agent Guide
Concise but detailed reference for contributors working on AUV. Improve code when you touch it; avoid one-off patterns and keep the shared runtime model in view.
Project Mission
- AUV turns application UI workflows into command-like, inspectable, replayable, and eventually shortcut-like operations.
- AUV is not only a CLI wrapper and not a generic LLM agent.
- Design around reusable typed operation APIs, first-party drivers, implicit run recording, artifact capture, replay, and inspection.
- Keep CLI, MCP, library calls, and future UI surfaces on the same execution model.
- Prefer explicit boundaries between operation execution, drivers, Rust command frontends, run storage, and reference documentation.
- Use
docs/TERMS_AND_CONCEPTS.mdas the shared vocabulary for run recording, inspection, trace data, artifacts, and viewer-facing APIs. - When a design introduces or changes a core term, update
docs/TERMS_AND_CONCEPTS.mdinstead of defining the term only inside a transient spec.
Many project details are still undecided. During design and implementation, communicate with users frequently and clearly to avoid misunderstandings, premature naming decisions, and avoidable rework.
Project Phase: Restore The AUV Core Lane
AUV is currently pulling its active roadmap back to the Application Use Via core: invoke, run recording, artifacts, inspection, app-local Rust commands, and distill/compile/run reuse across frontends. The former SkillBundle surface has been retired; do not reintroduce bundle execution, export, or verification as compatibility. The important work is to make the remaining execution surfaces agree on one shared execution model. Prefer changes that tighten or reconnect the existing core responsibilities over polishing one archived vertical proof.
There is no auv-runtime package or root Cargo package. Runtime is an
architectural responsibility distributed across typed operation modules,
frontend roots, drivers, and auv-tracing; do not recreate a catch-all runtime
crate without an owner-approved producer/consumer need.
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 · 5,903 tokens per session scan A 10d26876977a
auv AGENTS.md is an instructions file published in the GitHub repository moeru-ai/auv (25 stars, last pushed 5d ago), licensed Apache-2.0. It adds 5,903 tokens to every session, about $0.0295 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
activity-frames AGENTS.md
Instructions for nossa-y/activity-frames, covering can i install it without asking?, the mental model (10 seconds), fastest integration: mcp, from python and things you should not do.
clickproof CLAUDE.md
Claude Code instructions for sandeep-alluru/clickproof, covering clickproof — session anchor, key decisions, next step and mvp definition.
clickproof AGENTS.md
AGENTS.md instructions for sandeep-alluru/clickproof, covering clickproof — agent context, what this project does, module map, key invariants and testing.
clickproof copilot-instructions.md
Copilot instructions for sandeep-alluru/clickproof, covering github copilot instructions — clickproof, module map, key invariants, code style and adding a new output format.
kungfu copilot-instructions.md
Copilot instructions for kungfu-systems/kungfu: Read and follow AGENTS.md before proposing or running changes. It is the shared agent router for this repository, including the canonical Shifu development and build entrypoint.
amfs CLAUDE.md
Instructions for raia-live/amfs, covering amfs memory — agent instructions, available mcp tools, identity, brain tools (agent-scoped) and shared knowledge tools.