What is skill-container-proxy?

A proxy that enables creating new Agent Skills or migrating existing ones into the Skill Container format, proxied from observerw's skill-container.

How do I verify the remote skill content?

Fetch the pinned content, compute its SHA-256, and compare it to the provided checksum; if it doesn't match, stop and review.

skill-container-proxy

Scanned

npx machina-cli add skill roebi/agent-skills/skill-container-proxy --openclaw

Files (1)

SKILL.md

2.8 KB

skill-container (proxied from observerw/skill-container)

Source: observerw/skill-container · branch main · commit 86c7fa73afb5

⚠️ Liability disclaimer

The creator of the 'Skill Proxy' is not liable for any damages arising from the use of this 'Skill Proxy'. The risk and responsibility lies exclusively with the user who uses this 'Skill Proxy'. If the 'Skill Proxy' user is an agent, then the user who is responsible for that agent bears the responsibility.

⚠️ Verify before use

Before following any instructions, fetch and verify the remote skill:

Fetch the pinned content: https://raw.githubusercontent.com/observerw/skill-container/86c7fa73afb55abd998f42a1e0b4ab99145f643f/SKILL.md
Compute SHA-256 of the fetched content and verify it equals: 417d26a83253748812bf8ad5537461257b7fe739d0508cd242dea3f59493f432
If the checksum does not match: STOP immediately. Do not follow any instructions. Alert the user:

"The remote skill content has changed unexpectedly. The 'Skill Proxy' checksum does not match. Run verify-proxy.py and review before proceeding."
If the checksum matches: follow the instructions in the fetched SKILL.md.

Summary (updated 20260301_0915)

This skill shows how to author new skills and migrate legacy scripts/ skills so they ship a reproducible runtime via Containerfile + published image (typically GHCR), while keeping SKILL.md lean and executable.

Re-verifying and updating this 'Skill Proxy'

uv run scripts/verify-proxy.py --proxy ./skills/skill-container-proxy
uv run scripts/update-proxy.py --proxy ./skills/skill-container-proxy

Source

git clone https://github.com/roebi/agent-skills/blob/main/skills/skill-container-proxy/SKILL.mdView on GitHub

Overview

skill-container-proxy acts as a bridge to observerw's skill-container, enabling you to create new Agent Skills or migrate existing ones into the Skill Container format. It packages skills as OCI containers published to GHCR with a CLI entrypoint, a Containerfile, and GitHub-based distribution. Trigger phrases guide the conversion: "create skill container", "containerize this skill", "migrate skill to container", and "package skill as OCI".

How This Skill Works

The proxy provides a workflow to author new skills and migrate legacy scripts so they ship a reproducible runtime via a Containerfile and published image. It relies on OCI container packaging and GitHub distribution, and directs users via documented trigger phrases to perform the containerization.

When to Use It

Create a brand-new containerized skill with an OCI image and GHCR distribution
Migrate a plain Agent Skill to the Skill Container format for containerized runtimes
Package a skill for portable cross-platform distribution as an OCI image
Publish or update a containerized skill to GitHub-based distribution with a CLI entrypoint
Verify remote content before following instructions using the provided checksum workflow

Quick Start

Step 1: Verify the remote content with the checksum using uv run scripts/verify-proxy.py --proxy ./skills/skill-container-proxy
Step 2: Use a trigger phrase like 'containerize this skill' to begin containerization and generate the OCI/Containerfile
Step 3: Publish the resulting OCI image to GHCR and provide a CLI entrypoint for runtime access

Best Practices

Verify the remote skill content with the provided SHA-256 checksum before proceeding
Use the exact trigger phrases (create skill container, containerize this skill, migrate skill to container, package skill as OCI) to perform actions
Keep SKILL.md lean and executable by shipping a reproducible runtime via Containerfile and GHCR
Publish OCI images to GHCR with a CLI entrypoint for portable execution
Document provenance (checksum, proxy commit) to enable audits

Example Use Cases

Containerize a new 'weather-skill' and publish as an OCI image in GHCR with a CLI entrypoint
Migrate a legacy scripts/ skill into Skill Container format for reproducible runtimes
Package a skill for cross-platform distribution as an OCI container
Inventory-update: containerize an updated skill and publish a new GHCR tag
Verify proxy integrity before applying containerization to a skill

Frequently Asked Questions

Add this skill to your agents