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/altans/collie/claude-mdgit clone --depth 1 https://github.com/AltanS/collieWrote 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/altans/collie/claude-md)<a href="https://agentmods.dev/instructions/altans/collie/claude-md"><img src="https://agentmods.dev/badge/instructions/altans/collie/claude-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 | $0.09254 | $0.09254 |
| Opus 5 | $0.04627 | $0.04627 |
| Sonnet 5 | $0.01851 | $0.01851 |
| Haiku 4.5 | $0.00925 | $0.00925 |
Grade A, and why
collie CLAUDE.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 today.
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 — 447 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — working agreement for this repo
Collie (repo AltanS/collie) — a phone web UI for the AI agents running in your terminal,
served over Tailscale. A mobile-first PWA (Vite + React + TS + Tailwind v4 + shadcn) plus a Bun/TS
bridge that mirrors ONE multiplexer per install — Herdr, tmux or zellij — letting you monitor and
reply to agents from a phone. Herdr is one adapter among the three, not the product: it is the
default, it is the only one that talks over a Unix socket, and its plugin route stays supported —
plugin id herdr.collie (manifest: herdr-plugin.toml). Orientation:
README.md · ARCHITECTURE.md · the UI's visual
language DESIGN.md · verified API HERDR_API.md ·
decisions .adr/ · adding a harness
HARNESS_CONTRIBUTING.md · adding a multiplexer
MUX_CONTRIBUTING.md.
Decision records — read before reopening a settled question
.adr/ holds the decisions whose reasoning would otherwise live only in a PR thread —
specifically the ones that close off an option someone will reasonably propose again. If you're
about to argue why not rather than how, check there first; if the answer isn't there and the
decision is that shape, add one (numbering + format: .adr/README.md).
Rules elsewhere in this file stay short and normative and link to the ADR for the argument. Don't restate an ADR's reasoning here, and don't edit a superseded ADR into agreement with the present — mark it superseded and write the next one.
Versioning — MANDATORY
Collie is SemVered, and the version is enforced, so it never silently drifts.
The version lives in three files that must always agree, plus a matching CHANGELOG entry:
herdr-plugin.toml (canonical — Herdr reads it) · package.json · web/package.json ·
newest numbered ## [x.y.z] heading in CHANGELOG.md. ## [Unreleased] is not numbered and is
not part of that agreement.
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.
- today Changed · +221 lines · +4,662 tokens per session 6f26136ea891
- 4d ago First seen · 226 lines · 4,592 tokens per session scan A 9eeec0d9a11d
collie CLAUDE.md is an instructions file published in the GitHub repository AltanS/collie (739 stars, last pushed today), licensed MIT. It adds 9,254 tokens to every session, about $0.0463 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
herdr-agent-usage AGENTS.md
AGENTS.md instructions for senna-lang/herdr-agent-usage, covering project instructions, scope, provider extensibility, provider symmetry and directory and package boundaries.
herdr-sidebar CLAUDE.md
Instructions for alexarthurs/herdr-sidebar, covering herdr-sidebar monorepo, build / test / lint, plugin dev workflow, windows caveats (verified against herdr 0.7.1 and 0.8.x) and release flow (verified for v0.7.0).
herdr-cache-alert CLAUDE.md
Instructions for AltanS/herdr-cache-alert, covering claude.md — working agreement for this repo, the claim contract — mandatory, versioning — mandatory, build / run and architecture — the four layers.
herdr-extensions CLAUDE.md
Instructions for vonzelle-vzt/herdr-extensions, covering claude.md — herdr-extensions, what this is, repository map, 🔴 the five things that fail silently and conventions.
herdr-agent-watcher AGENTS.md
AGENTS.md instructions for winoooops/herdr-agent-watcher, covering agents.md — agent watcher, project overview, runtime architecture, module map and build and test commands.
herdr-agent-watcher CLAUDE.md
Claude Code instructions for winoooops/herdr-agent-watcher, covering claude.md, project summary, common commands, architecture and module map.