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/openvang/vanguarstew/agents-mdgit clone --depth 1 https://github.com/openvang/vanguarstewWrote 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/openvang/vanguarstew/agents-md)<a href="https://agentmods.dev/instructions/openvang/vanguarstew/agents-md"><img src="https://agentmods.dev/badge/instructions/openvang/vanguarstew/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 | $0.00420 | $0.00420 |
| Opus 5 | $0.00210 | $0.00210 |
| Sonnet 5 | $0.00084 | $0.00084 |
| Haiku 4.5 | $0.00042 | $0.00042 |
Grade A, and why
vanguarstew 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 4d 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 — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.
OpenVang project constitution
Durable project-wide rules for the OpenVang agent factory and the Vanguarstew maintainer-intelligence component. Code, tests, and automation must follow these rules.
Component contract
- The maintainer component shall expose
solve(repo_path, request, ...)as its stable entrypoint. - Managed inference parameters are supplied by the controller; agent code shall not discover or substitute credentials.
VANGUARSTEW_OFFLINE=1shall select the deterministic offline stub.- Benchmark and live persistent memory shall remain time-safe, controller-owned, and read-only from the maintainer component.
Benchmark and execution integrity
- Forward-looking signal shall not cross a benchmark freeze boundary.
- Held-out repositories shall be evaluated separately from tuned repositories.
- Public artifacts and TEE evidence shall contain only receipt-safe commitments, never raw private memory, review material, credentials, or private sources.
- Polaris integration shall be described as execution integrity, not workload confidentiality.
Factory authority
- Every factory worker shall declare one role contract from
openvang/factory.py. - No role may automatically access a wallet, submit an on-chain transaction, change emissions, vote in governance, mutate GitHub, or publish.
- Owner-level effects require a separate external approval and signing boundary;
a factory
ActionIntentis non-executable by design. - Role-private memory, including private maintainer-review material, shall never cross role boundaries or enter public, benchmark, or TEE artifacts.
- Security QA is defensive and isolated; it may propose containment but may not perform an offensive or production mutation.
Quality gates
- Changes under
agent/,benchmark/,openvang/, orvanguarstew_runtime/shall include matching tests. ruff check .and the relevant offline test suite shall pass before release.- Runtime defaults shall remain dry-run, loopback-only, private, and without a GitHub write path.
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.
- 4d ago First seen · 45 lines · 420 tokens per session scan A 00ad6fb86748
vanguarstew AGENTS.md is an instructions file published in the GitHub repository openvang/vanguarstew (21 stars, last pushed yesterday), licensed MIT. It adds 420 tokens to every session, about $0.0021 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
mem0 AGENTS.md
AGENTS.md instructions for mem0ai/mem0, covering agents.md, do not, where to look, repository map and setup.
pydantic-ai AGENTS.md
AGENTS.md instructions for pydantic/pydantic-ai, covering your primary responsibility is to the project and its users, gathering context on the task, ensuring the task is ready for implementation, philosophy and requirements of all contributions.
parlant CLAUDE.md
Instructions for emcie-co/parlant: This is the main repo of Parlant (https://parlant.io).
redis-vl-python CLAUDE.md
Claude Code instructions for redis/redis-vl-python, covering claude.md - redisvl project context, frequently used commands, development workflow, redis setup and documentation.
strix-halo-guide AGENTS.md
AGENTS.md instructions for hogeheer499-commits/strix-halo-guide, covering agents.md, core project lens, documentation style, do-not-invent rules and benchmark claim rules.
openbench AGENTS.md
Instructions for minghinmatthewlam/openbench, covering openbench — agent context, local execution context, what openbench is, execution ownership and product goals (the two things we are building toward).