Run Python and shell code safely inside Hermes agent tasks
Runs code in a sandboxed environment. Review permissions before use in production.
Allows your Hermes agent to write and execute Python or shell scripts in a sandboxed environment. Useful for data processing, automation scripts, and dynamic calculations.
Read this skill's SKILL.md in full before installing it. Run it against a disposable host or test account first, and give it the narrowest credentials that still work.
The listing carries an explicit maintainer warning. Test this against something disposable before pointing it at anything you care about.
Listing carries a maintainer warning: "Runs code in a sandboxed environment. Review permissions before use in production."
This skill executes commands on the machine running the agent. That is the broadest permission you can grant: anything the agent user can do, this skill can do.
This skill's own description refers to "shell"
It works with files on disk. Consider which directories the agent process can actually reach.
This skill's own description refers to "file"
These permissions are inferred from this listing — the credentials it declares, any maintainer warning, and the operations its own description mentions. ClawVault has not audited this skill's source code. Treat it as a starting point for your own review, not a substitute for one.
Hermes skills are on-demand knowledge documents stored in hermes/skills/, which is the single source of truth. Bundled skills are copied there on a fresh install, and hub-installed and agent-created skills land there too. Every installed skill automatically becomes a slash command, so a skill named plan is invoked with /plan.
hermes skills search code-execAdd --source to widen the search: official for curated skills, skills-sh for the public skills.sh directory, browse-sh for site-specific browser automation, or well-known to point Hermes at a site publishing its own skill index.
hermes skills inspect official/<category>/code-execinspect shows the skill's contents and metadata. Use the fully qualified identifier that search returned, which includes its source and category.
hermes skills install official/<category>/code-execIdentifiers are source-qualified, for example official/migration/openclaw-migration or skills-sh/anthropics/skills/pdf. You can also install straight from a URL that serves a SKILL.md. --force overrides non-dangerous policy blocks but will not override a dangerous scan verdict.
/code-execInstalled skills become slash commands in any chat session. Run hermes skills list to see everything available, or /skills from inside a session.
Commands follow the official Hermes Agent documentation. Placeholders such as <owner> are filled in by the search step, because registry references are publisher-scoped.
Hermes checks third-party and community skills against policy before installing, and --force will not override a verdict it considers dangerous. Because Hermes can also write its own skills at runtime, review what the Curator has kept periodically with hermes skills list.
Read this skill's SKILL.md before installing it, so you know what commands it runs and which credentials it reads.
Deploy OpenClaw, Hermes, or PicoClaw on Bluehost with the right template for the job.
The GitHub skill gives your OpenClaw agent complete access to GitHub — creating and reviewing pull requests, managing issues, listing repositories, checking build status, browsing code and triggering workflows. For developers running OpenClaw as a coding assistant this skill is essential. It closes the loop between your agent writing code and that code actually landing in a repository. Widely used in automated PR workflows, issue triage pipelines and code review automations. Works seamlessly alongside Composio for teams that need multi-service integration.
N8N Workflow connects your OpenClaw agent to your N8N automation instance, letting the agent trigger workflows, pass data between systems and monitor automation runs through plain conversation. If you already use N8N for business automation, this skill turns your agent into the intelligent orchestration layer on top — deciding which workflows to trigger based on context rather than fixed schedules. Particularly powerful for teams using N8N as their backend automation layer who want to add AI decision-making on top.
Claude Vision enables agents to understand and analyze images, screenshots, diagrams, and visual content. Perfect for agents that process visual information, extract data from images, or analyze screenshots and diagrams.