Opencode Omo
PassAudited by ClawScan on May 1, 2026.
Overview
This instruction-only skill coherently guides OpenCode coding workflows; the main cautions are to review agent-made code changes and not run unprovided helper scripts blindly.
This appears safe to install as a workflow guide. Use it in a git-controlled project, review any generated code before accepting it, and do not run referenced helper scripts unless you have inspected their source.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
A poorly scoped request could lead to unintended code changes or implementation choices.
The skill directs the user to run an agentic coding runtime that can implement changes in a project. This is central to the skill's purpose, but it is still a powerful workflow.
opencode run --agent sisyphus "ulw implement JWT auth in this service and add tests" ... Use `/start-work` to let Atlas execute the plan.
Run it in the intended repository, keep work under version control, and review the plan, diffs, and test results before accepting or committing changes.
If a user tries these helper commands, they may fail or run a local script that was not part of the reviewed skill artifact.
The reviewed file manifest contains only SKILL.md, so the referenced helper scripts are not included for review. The examples are user-directed, but their provenance should be verified before running.
You can also run the bundled checker: `./scripts/check-omo.sh` ... Or use the bundled helper: `./scripts/run-ulw.sh`
Prefer the explicit OpenCode commands, or inspect and verify any `./scripts` helper before executing it.