Stakpak is an open-source DevOps agent that runs continuously on a machine to keep applications running and automate infrastructure work. It helps teams generate infrastructure code, troubleshoot Kubernetes, configure CI/CD, and automate deployments while limiting an LLM’s access to production. The catalogue add-ons provide commands and instructions for operating Stakpak.
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 commands/stakpak/agent/remote-autopilot-install.v1git clone --depth 1 https://github.com/stakpak/agentWrote 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/commands/stakpak/agent/remote-autopilot-install.v1)<a href="https://agentmods.dev/commands/stakpak/agent/remote-autopilot-install.v1"><img src="https://agentmods.dev/badge/commands/stakpak/agent/remote-autopilot-install.v1.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 | $0.00036 | $0.06214 |
| Opus 5 | $0.00018 | $0.03107 |
| Sonnet 5 | $0.00007 | $0.01243 |
| Haiku 4.5 | $0.00004 | $0.00621 |
Grade B, and why
remote-autopilot-install.v1 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.
command -v sudo >/dev/null && echo HAS_SUDO || true; \ Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -sSL https://raw.githubusercontent.com/noureldin-azzab/stakpak-autopilot-install/06de6725a85301d6bc026ad5a400f60e6974c477/autopilot-install.sh | \ How it starts
The opening of the file, as written. The whole thing — 906 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an installer agent. Install Stakpak Autopilot on a remote Linux host by gathering all required inputs with ask_user, resolving a reachable SSH target, then passing those values to the generic bootstrap script in one remote execution.
The bootstrap script is vendor-neutral Linux. Keep cloud-provider API calls in this skill as target adapters. Target adapters discover or provision a VM, confirm SSH reachability, and return a normalized SSH target.
Input: {input}
Target outcome
A remote Linux host has:
- Stakpak CLI installed
- Docker installed and usable by the target user
- Autopilot running as a systemd user service
- Auth profile configured with a valid model
- Notification channel configured, unless skipped
- Notification routing configured to a destination such as
#prod, a chat ID, or a channel ID stakpak autopilot statushealthy
Expected runtime:
- Existing Linux VM with Docker already present: ~30–60s
- Existing fresh Linux VM: ~90–120s
- Full cloud VM provisioning plus install: usually ~3–5 minutes, provider-dependent
Architecture rule
Keep the workflow split into two layers:
| Layer | Responsibility | Vendor-specific? |
|---|---|---|
| Target adapter | Find or provision a reachable Linux host and collect SSH details | Yes |
| Bootstrap script | Install/configure Stakpak Autopilot on that host | Provider-neutral |
Every target adapter must return the same normalized output:
ssh_target=<user>@<host>[:port]
ssh_credential_method=default-keys|private-key|password
target_user=<linux-user>
private_key_path=<optional-local-path>
password=<optional-password>
provider=<ssh|aws|gcp|azure>
source_resource_id=<optional-provider-resource-id>
public_ip=<optional-public-ip>
private_ip=<optional-private-ip>
After an adapter returns this data, all providers use the same probe, install, and verify phases.
Input collection pattern
Use ask_user as the primary UX mechanism for collecting choices and sensitive values.
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 · 906 lines · 36 tokens per session scan B 775d48e4fc8c
remote-autopilot-install.v1 is a command published in the GitHub repository stakpak/agent (1,768 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 36 tokens to every session and 6,214 once invoked, about $0.0002 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 commands, from other repositories
b00t
Show b00t hive status — git, zellij, governance gates, cake balance, open issues.
deploy-ingen-test
Deploy SoCa or Prompt Tuner applications to Azure using the ingen-test resource group.
az-daily-summary
Generate a comprehensive daily summary of Azure subscription activity using Azure CLI.
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.