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/sous-chefs/varnish/agents-mdgit clone --depth 1 https://github.com/sous-chefs/varnishWrote 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/sous-chefs/varnish/agents-md)<a href="https://agentmods.dev/instructions/sous-chefs/varnish/agents-md"><img src="https://agentmods.dev/badge/instructions/sous-chefs/varnish/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 | $0.00886 | $0.00886 |
| Opus 5 | $0.00443 | $0.00443 |
| Sonnet 5 | $0.00177 | $0.00177 |
| Haiku 4.5 | $0.00089 | $0.00089 |
Grade A, and why
varnish 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 — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Agent Guidance
This cookbook uses Policyfile.rb for dependency resolution. Run chef install Policyfile.rb
after changing cookbook dependencies or test cookbook recipes, and keep Kitchen suites mapped to
Policyfile named run lists.
Package Availability
APT (Debian/Ubuntu)
The Varnish Cache project hosts official community packages on Packagecloud and recommends manual repository setup rather than the Packagecloud installer scripts.
- Ubuntu 20.04, 22.04, and 24.04 are documented by Varnish Software for Varnish Cache 6.0 LTS packages.
- Debian and Ubuntu distro packages are available through the native package manager, but distro versions may differ from Varnish Cache LTS.
- Packagecloud repository URLs use the pattern
https://packagecloud.io/varnishcache/varnish<major_without_dot>/<platform>/. - During local Dokken verification on May 7, 2026,
https://packagecloud.io/varnishcache/varnish60/ubuntu/ noblereturned 404 for the Ubuntu 24.04 arm64 container, so the required default suite uses distro packages.
DNF/YUM (RHEL family)
The official Varnish documentation describes Packagecloud RPM repositories for Enterprise Linux. Varnish Cache 6.x package documentation explicitly covers EL 7 and EL 8. EL 9 compatible platforms may have distro packages, but this cookbook only treats Packagecloud availability as source-verified where upstream documents it.
- RHEL-family platforms need the stock
varnishDNF module disabled on EL 8 and newer before installing Packagecloud packages. - EPEL is still included for dependency compatibility on RHEL-family systems.
- Packagecloud repository URLs use the pattern
https://packagecloud.io/varnishcache/varnish<major_without_dot>/el/<major>/$basearch.
Zypper (SUSE)
No source-verified community Packagecloud support for openSUSE Leap was found for this cookbook migration, so openSUSE is not listed in metadata.rb or Kitchen.
Architecture Limitations
Packagecloud repository paths are architecture-aware for RPMs through $basearch. The Varnish documentation consulted does not provide a complete per-release architecture matrix for community DEB/RPM repositories, so the cookbook does not claim architecture-specific support beyond what the upstream repositories publish.
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 · 51 lines · 886 tokens per session scan A e5083ed7a648
varnish AGENTS.md is an instructions file published in the GitHub repository sous-chefs/varnish (55 stars, last pushed 18d ago), licensed Apache-2.0. It adds 886 tokens to every session, about $0.0044 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
postfix copilot-instructions.md
Instructions for sous-chefs/postfix, covering copilot instructions for sous chefs cookbooks, repository overview, project structure, build and test system and environment setup.
postfix AGENTS.md
Instructions for sous-chefs/postfix: This cookbook manages Postfix through operating system packages and service/configuration files. It does not build Postfix from source.
ossec AGENTS.md
Instructions for sous-chefs/ossec, covering agents.md, cookbook purpose, agent findings, known limitations and package availability.
nfs AGENTS.md
Instructions for sous-chefs/nfs, covering agent notes, package availability, apt (debian/ubuntu), dnf/yum (rhel family, fedora, amazon linux) and zypper (suse).
emacs AGENTS.md
Instructions for sous-chefs/emacs, covering agents.md, cookbook purpose, agent findings, known limitations and package availability.
beyondcompare AGENTS.md
Instructions for sous-chefs/beyondcompare, covering agents.md, cookbook purpose, agent findings, known limitations and package availability.