Hormuzmonitorskills

WarnAudited by ClawScan on May 10, 2026.

Overview

The skill’s monitoring purpose is coherent, but it asks the agent to automatically run an unreviewed deployment script that pushes updates to GitHub/Vercel without clear approval or credential boundaries.

Before installing, inspect ~/hormuz-website/scripts/update_and_deploy.sh, confirm the exact GitHub repository and Vercel project it can modify, and require manual approval before any push or deployment.

Findings (4)

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.

What this means

The agent could change local files and trigger a public website update whenever the skill runs.

Why it was flagged

The skill instructs the agent to execute a shell command automatically after each cycle, without requiring user approval or constraining what the script may do.

Skill content
After writing MONITOR_LOG.md, run this command to update the website:

exec command: bash ~/hormuz-website/scripts/update_and_deploy.sh
Recommendation

Require explicit user approval before running the deployment script, document exactly what the script changes, and consider a dry-run or staging workflow.

What this means

A user cannot verify from these artifacts what code will run or what repository, credentials, or files it will affect.

Why it was flagged

The referenced helper script is outside the supplied skill artifacts, and there is no install spec or reviewed code showing its contents or provenance.

Skill content
exec command: bash ~/hormuz-website/scripts/update_and_deploy.sh
Recommendation

Include the deployment script in the reviewed artifact set or require users to inspect and approve the local script path before enabling this step.

What this means

The agent may use the user’s existing GitHub/Vercel publishing privileges to modify a public site.

Why it was flagged

Pushing to GitHub uses delegated account or repository authority, but the skill does not declare or bound which credentials, repo, branch, or permissions are used.

Skill content
This parses MONITOR_LOG.md into data.json and pushes to GitHub.
Vercel auto-deploys on push.
Recommendation

Declare the exact repository and credential requirements, use least-privilege credentials, and require confirmation before any push.

What this means

Bad source data or a formatting mistake could be published to the website and persist until the next successful update.

Why it was flagged

A monitoring/logging output is automatically propagated through GitHub into a live Vercel deployment, so incorrect or manipulated data could become public without a review gate.

Skill content
This parses MONITOR_LOG.md into data.json and pushes to GitHub.
Vercel auto-deploys on push.
Recommendation

Add a manual review or staging step before production deployment, and document rollback procedures.