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/newsnowlabs/dockside/claude-mdgit clone --depth 1 https://github.com/newsnowlabs/docksideWrote 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/newsnowlabs/dockside/claude-md)<a href="https://agentmods.dev/instructions/newsnowlabs/dockside/claude-md"><img src="https://agentmods.dev/badge/instructions/newsnowlabs/dockside/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 5.1 | $0.02593 | $0.02593 |
| Opus 5 | $0.01296 | $0.01296 |
| Sonnet 5 | $0.00519 | $0.00519 |
| Haiku 4.5 | $0.00259 | $0.00259 |
Grade B, and why
dockside CLAUDE.md scanned grade B with 2 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 6d 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 s6-svc -t /etc/service/nginx` Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
`curl` — it reuses the CLI's own already-working session/TLS/connection setup. Confirmed this How it starts
The opening of the file, as written. The whole thing — 166 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Repo Notes
./test.shruns the full static suite (Perl compile, Vue build, ESLint, StyleLint, ShellCheck, JSON/YAML, Python compile) — run it regularly, not just for Perl../test.sh --only <check>runs a single check (e.g.--only perlwhile iterating on Perl underapp/server/liborapp/server/bin;--only vue,--only eslint, …).- When validating Perl server changes on this local Dockside host, restart:
sudo s6-svc -t /etc/service/nginxsudo s6-svc -t /etc/service/docker-event-daemon
- Before running tests that exercise server changes, restart the services above —
the running server is not auto-reloaded. Rebuild the Vue bundle
(
cd app/client && npm run build) for client changes.
The dockside CLI
Needed for two things: running the integration suite, and ad-hoc HTTP checks against a live instance mid-session (both below). Nothing else in this repo — static checks, restarting services, editing code — needs a CLI session at all.
Session setup
Check for an existing session before assuming there isn't one or that you need
credentials: run python3 cli/dockside server list (or dockside server list if it's on
PATH). Recommended usage is for the developer to already be logged in as admin locally and
hand that session to you, so one usually exists — never ask the user for credentials over
chat. If none is found, ask the user whether they'd rather log in themselves (remind them of
the login command below) or have you try it — don't just assume either way.
If asked to try it yourself: the hostname depends on your environment:
- Personal laptop/server running Dockside:
www-<name>.local.dockside.dev - Inner Dockside dev container inside an outer production Dockside: your inner
container's public hostname (e.g.
www-<name>.dockside-domain.com— the outer server's domain is whatever it actually is, no fixed convention to assume) — derive it rather than asking or guessing:ssl.domains[0]in/data/config/config.jsongives the base domain (e.g.<name>.dockside-domain.com); prependwww-to its first label for the actual UI hostname (www-<name>.dockside-domain.com). Don't trust the container's own "Navigate to ..." boot-log line for this — confirmed unreliable: on a real instance it printed awww.<name>...form (dot, not hyphen) that 400s, while the actual working hostname was the hyphenatedwww-<name>...form above.
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.
- 6d ago First seen · 166 lines · 2,593 tokens per session scan B 1a4681ce8414
dockside CLAUDE.md is an instructions file published in the GitHub repository newsnowlabs/dockside (322 stars, last pushed 2d ago), licensed Apache-2.0. It adds 2,593 tokens to every session, about $0.0130 per session on Opus 5. A static security scan graded it B with 2 findings (asks for root, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
netrock CLAUDE.md
Claude Code instructions for fpindej/netrock, covering claude.md, hard rules, backend, frontend and cross-cutting.
factory-factory AGENTS.md
Instructions for purplefish-ai/factory-factory, covering factory factory — agent guide, everyday commands, before you hand work back, layout and architecture rules.
factory-factory CLAUDE.md
Instructions for purplefish-ai/factory-factory, a project described as: Workspace-based coding environment for running multiple Claude Code and Codex sessions in parallel.
appwrite AGENTS.md
AGENTS.md instructions for appwrite/appwrite, covering appwrite, commands, stack, layout and libraries.
aider-desk AGENTS.md
AGENTS.md instructions for hotovo/aider-desk, covering agents.md, common commands, development, type checking and linting and formatting.
mcp-dockhand copilot-review.instructions.md
Copilot instructions for strausmann/mcp-dockhand: When reviewing pull requests in this repository.