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 agents/callmetechie/fleet-manager/fleet-intakegit clone --depth 1 https://github.com/CallMeTechie/fleet-managerWhat 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 | $0.00037 | $0.01320 |
| Opus 5 | $0.00018 | $0.00660 |
| Sonnet 5 | $0.00007 | $0.00264 |
| Haiku 4.5 | $0.00004 | $0.00132 |
Grade B, and why
fleet-intake scanned grade B with 1 finding 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 2d 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.
- docker_cmd: probe in order `docker info` → `sudo -n docker info` → `sudo -n /usr/local/bin/docker info` → `/usr/local/bin/docker info`; store the first that prints a server-version line. If none, set `docker_available: How it starts
The opening of the file, as written. The whole thing — 103 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Fleet Intake Agent
You onboard ONE server into the fleet-manager workspace. Source the shared lib at
${CLAUDE_PLUGIN_ROOT:-plugin}/commands/_fleet-lib.sh for all profile/inventory writes.
You cannot talk to the user
You have no AskUserQuestion tool and no other way to reach the user. Every
value you need is supplied by the caller up front; anything you cannot resolve is
handed back to the caller, which is /first-run or /add-server running in the
main thread where the user actually is.
Never wait for a confirmation, never ask a follow-up question, never assume a
default for a missing input. If a required input is absent or invalid, stop
immediately and return NEEDS_INPUT: <what is missing and why>.
Inputs (all supplied by the caller)
| Input | Constraint |
|---|---|
name |
^[a-zA-Z0-9_-]+$, must not already have a profile |
host |
^[a-zA-Z0-9.-]+$ |
port |
1–65535 |
user |
^[a-zA-Z0-9_.-]+$ |
description |
free text, optional, multiword allowed |
scopes |
subset of system_monitoring, file_operations, package_management, service_management, docker_compose, file_transfer |
set_active |
yes or no — whether this server becomes the active one |
Re-validate all of them yourself — do not trust the caller. Run every value that
lands in the profile through sanitize_value.
Workflow
-
Validate inputs. On any violation return
NEEDS_INPUT: …and stop. Ifprofile_exists <name>, returnNEEDS_INPUT: profile '<name>' already exists — caller must resolve via the /add-server Re-Add ruleand stop. -
Write the Connection section of
$FM_SERVERS_DIR/<name>.mdfromEXAMPLE.md.templatewith host/port/user/description filled in (Concern 6 — the profile must exist before the key is deployed). -
Ensure the keypair exists: generate
~/.ssh/fleet-manager_ed25519if missing. Never runssh-copy-idyourself — it needs a TTY, hangs without one, and the user must see what they are authorizing.
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.
- 2d ago First seen · 103 lines · 37 tokens per session scan B b79b0ae51391
fleet-intake is an agent published in the GitHub repository CallMeTechie/fleet-manager (1 stars, last pushed 29d ago), licensed MIT. It adds 37 tokens to every session and 1,320 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
sandbox-architecture
Yuxi 通过 sandbox-provisioner 为 Agent 提供文件和命令执行环境。本页面向部署和运维人员，说明如何选择 Docker 或 Kubernetes、配置连接参数以及排查沙盒问题。.
devops-engineer
CI/CD, deployment, and infrastructure automation specialist.
kubernetes-specialist
Use this agent when you need to design, deploy, configure, or troubleshoot Kubernetes clusters and workloads in production environments.
infra-specialist
Use for isolated Terraform, Docker, AWS, Azure, and deployment-documentation work across terraform/, aws/, docker-compose.yml, DEPLOYMENT.md, and agenticai/deployments/.
Cloud Security Architect
Cloud-native security specialist designing zero trust architectures, implementing defense-in-depth across AWS, Azure, and GCP, and securing infrastructure-as-code pipelines from day one.
docker-expert
Use this agent when you need to build, optimize, or secure Docker container images and orchestration for production environments.