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 skills add Gentech-Labs/genTech-agent-kit --skill fleet-update-containergit clone --depth 1 https://github.com/Gentech-Labs/genTech-agent-kitWrote 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/skills/gentech-labs/gentech-agent-kit/fleet-update-container)<a href="https://agentmods.dev/skills/gentech-labs/gentech-agent-kit/fleet-update-container"><img src="https://agentmods.dev/badge/skills/gentech-labs/gentech-agent-kit/fleet-update-container/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/gentech-labs/gentech-agent-kit/fleet-update-container"><img src="https://agentmods.dev/badge/skills/gentech-labs/gentech-agent-kit/fleet-update-container.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00099 | $0.02333 |
| Opus 5.5 | $0.00040 | $0.00933 |
| Sonnet 5.5 | $0.00020 | $0.00467 |
| Haiku 4.5 | $0.00010 | $0.00233 |
Grade A, and why
fleet-update-container 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 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.
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 — 186 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Fleet Update — Container Edition
When to Use
- Your agent runs as a Docker container, not a bare-metal install — you have no writable
/opt/hermes, nosystemd, and/opt/datais a host bind mount. - You are about to follow
safe-update-restartand the commands look wrong for your box. - You want to know whether a newer image even exists before touching anything.
Read
safe-update-restartfirst — then come here. That skill owns the general restart doctrine. This one owns the container-specific facts, and several of that skill's commands do not work in a container. If your box is bare metal (you cancd /usr/local/lib/hermes-agentand it has a.git), usesafe-update-restartand stop here.
The one thing to understand first
In a container, the engine IS the image. On bare metal a restart reloads a process. In a container an update is a container recreate — a new image is pulled and the container is rebuilt on it. That changes the whole shape:
- Your state survives by design — it lives on a bind mount (
/opt/data), which the recreate does NOT touch./is the overlay layer that is replaced. - The engine cannot be patched —
/opt/hermes/.venvis baked into the image. A new engine means a new image, always. - The update is a HOST action — an agent inside the container cannot do it. It cannot reach the
Docker socket (uid 10000, denied), cannot write the install tree, and
hermes updaterefuses outright (published image, no.git, not writable).
Five assumptions that break in a container
| # | The bare-metal command | What actually happens in a container |
|---|---|---|
| 1 | hermes update --yes |
Refuses. Published image — no working tree to pull into. Not a permissions bug. |
| 2 | cd /usr/local/lib/hermes-agent |
Doesn't exist. Install is /opt/hermes, read-only to you. |
| 3 | systemctl is-active ... |
No systemd. Services are s6: s6-svstat /run/service/<name>, list with ls /run/service. |
| 4 | state in ~/.hermes, backup to /root/... |
/root is unreadable; state is the bind mount /opt/data. |
| 5 | "run the update" is your action | It's the host operator's action — docker compose up -d --force-recreate. |
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 · 186 lines · 99 tokens per session scan A 30edd8faeb26
fleet-update-container is a skill published in the GitHub repository Gentech-Labs/genTech-agent-kit (0 stars, last pushed 3d ago), licensed MIT. It adds 99 tokens to every session and 2,333 once invoked, about $0.0004 per session on Opus 5.5. 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 2026-10-22.
Other skills, from other repositories
bazel-oci-expert
Expert knowledge for building OCI container images with Bazel using rulesoci. Use for containerizing Quarkus/Java applications, multi-arch builds, and image optimization.
DevOps & Deployment
CI/CD pipelines, containerization, Kubernetes, and infrastructure as code patterns.
buildah
Buildah OCI container image builder reference. Build container images without Docker daemon. Covers Containerfile builds, scripted builds with shell commands, direct filesystem mount, rootless operation, multi-arch manifests, and CI/CD integration.
docker-registry
Container image registry workflows — GHCR, Docker Hub, and private registry auth, tagging strategies, CI push pipelines, image pruning, and multi-platform manifest publishing.
ecspresso
ECS deployment tool - deploy, manage, and troubleshoot ECS services.
04-security
A security and operations guide for deploying Dify, an open-source platform for building AI applications, in production.