factory-infra-scout

factory-infra-scout is an agent for Claude Code from watt-mind/factory. It costs 116 tokens per session (1,125 once invoked), scanned B, original, Apache-2.0.

A read-only investigator for deployed applications and infrastructure. It checks servers, containers, databases, DNS, deployment systems, and health endpoints to answer one operational question.

In plain words
What is it for?
Use it to check whether a development stack is healthy, inspect deployed environment values, confirm that a deployment used a new image, or investigate why a smoke test failed.
Why use it?
It avoids making developers sift through SSH sessions, container logs, and deployment details themselves. It also uses the project’s infrastructure documentation to find the relevant systems and settings.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: model in frontmatter; names the NotebookEdit tool.

Part of the core plugin — 2 skills, 13 commands, 4 agents shipped together

Install

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.

agentmods
npx agentmods add agents/watt-mind/factory/factory-infra-scout
Clone the repo
git clone --depth 1 https://github.com/watt-mind/factory

Made for: Claude Code.

Or install core, the plugin that ships this one along with the rest of its 2 skills, 13 commands, 4 agents.

Wrote 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.

agentmods badge for factory-infra-scout

README.md
[![agentmods](https://agentmods.dev/badge/agents/watt-mind/factory/factory-infra-scout.svg)](https://agentmods.dev/agents/watt-mind/factory/factory-infra-scout)
Your own site
<a href="https://agentmods.dev/agents/watt-mind/factory/factory-infra-scout"><img src="https://agentmods.dev/badge/agents/watt-mind/factory/factory-infra-scout.svg" alt="Measured on agentmods" height="20"></a>
Per session 116 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,125 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 2 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00116 $0.01125
Opus 5 $0.00058 $0.00562
Sonnet 5 $0.00023 $0.00225
Haiku 4.5 $0.00012 $0.00112

Measured 6d ago against content hash 4025fa714189, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade B, and why

factory-infra-scout 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 docker compose -p <project> ps / logs / top`, `docker inspect`, `docker stats --no-stream` — the SSH user typically is not in the `docker` group, so `sudo` is expected.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

- Hit health endpoints and public URLs (`curl -sS -o /dev/null -w '%{http_code}'`), check DNS, check certificate expiry.
plugins/core/agents/factory-infra-scout.md · 62 lines

How it starts

The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an infrastructure scout. You answer one question about deployed systems by investigating them directly, and you return a verdict with minimal evidence — never the raw output you waded through. Your entire value is that the caller's context stays clean: a docker compose logs tail is cheap for you and poison for them.

Inputs you should expect in your prompt

  • One question, as an outcome ("is bj29-dev healthy?", "what DATABASE_URL is the coach-wattz dev container actually running with?", "did PR #84's deploy land?").
  • The app and/or server name, if the caller knows it.

That is all you need. Everything else you resolve yourself from the source of truth:

  • ~/Develop/hdkiller/docs/servers/<name>.md — SSH access (prefer the Tailscale IP), credentials, specs.
  • ~/Develop/hdkiller/docs/applications/<app>.md — Dokploy resource IDs, compose project names, env vars, DB connection details, domains.
  • ~/Develop/hdkiller/docs/guides/dokploy-deployment.md — the Dokploy workflow, DNS, known troubleshooting (including compose-project-name collisions).

Read the relevant doc before SSHing. Do not ask the caller for hostnames, ports, or credentials that these docs contain; if a doc is missing or stale, that is itself a finding.

What you may do

Freely (read-only):

  • SSH into servers (ssh hdkiller@<tailscale-ip>), run read-only commands.
  • sudo docker compose -p <project> ps / logs / top, docker inspect, docker stats --no-stream — the SSH user typically is not in the docker group, so sudo is expected.
  • Read env vars off running containers (docker exec <c> env, docker inspect).
  • Hit health endpoints and public URLs (curl -sS -o /dev/null -w '%{http_code}'), check DNS, check certificate expiry.
  • Check what image/commit a container runs versus what the repo/registry says it should.
  • Read Dokploy state where the docs give you access.

Allowed, but only when the task explicitly asks for it — and always reported in your verdict:

Read the full file on GitHub · 62 lines

Changes

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.

  1. 6d ago First seen · 62 lines · 116 tokens per session scan B 4025fa714189

Subscribe to this mod's changes

factory-infra-scout is an agent published in the GitHub repository watt-mind/factory (14 stars, last pushed today), licensed Apache-2.0. It adds 116 tokens to every session and 1,125 once invoked, about $0.0006 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 2026-08-31.