Afrexai Business Automation TEMP

ReviewAudited by ClawScan on May 10, 2026.

Overview

This is mostly a documentation-only automation framework, but its package identity is inconsistent and it encourages broad automations that could change finance, HR, and account systems without clear approval boundaries.

Before installing, confirm the intended package name and publisher. Treat this skill as a planning aid unless you explicitly approve implementation steps. If you use it to build real automations, use least-privilege credentials, sandbox tests, human approval for finance/HR/account changes, monitoring, and documented rollback steps.

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

A user may be reviewing one package but be directed toward a differently named package or unclear owner, making it harder to verify provenance.

Why it was flagged

The registry identity differs from the embedded _meta.json slug/owner and the README install command, which point to afrexai-business-automation rather than the evaluated temp package.

Skill content
Slug: afrexai-business-automation-temp; Owner ID: kn7favg91765ttv6tt0fb1ycv582wktt
Recommendation

Verify the publisher and intended slug before installation, and ask the maintainer to align registry metadata, _meta.json, and README install instructions.

ConcernMedium Confidence
ASI02: Tool Misuse and Exploitation
What this means

The agent could help build or run automations that modify business records, send notifications, or operate recurring jobs without a clearly stated approval gate in the artifacts.

Why it was flagged

This directs the agent toward broad execution and integration mechanisms; combined with finance, HR, CRM, and support automation examples, this can lead to high-impact changes if not explicitly approved and scoped.

Skill content
Implement using agent tools — cron jobs, APIs, scripts, notifications
Recommendation

Use the skill for design and documentation by default, and require explicit human approval before creating scripts, cron jobs, API integrations, payments, account changes, or production workflow changes.

What this means

If connected to real admin tools, the agent could affect employee accounts, team memberships, and business access controls.

Why it was flagged

The onboarding workflow examples involve privileged account administration, which is expected for business automation but sensitive.

Skill content
→ Create accounts (email, Slack, GitHub, etc.)
  → Add to teams & channels
Recommendation

Only connect least-privilege admin credentials, test in a sandbox first, and require approval for account creation, permission changes, and access revocation.

NoteHigh Confidence
ASI10: Rogue Agents
What this means

Scheduled automations can continue running and affecting systems after setup if they are not tracked and removable.

Why it was flagged

The workflow template includes scheduled execution, which is normal for automation but can persist beyond the initial request.

Skill content
cron: "0 9 * * 1-5"  # Weekdays at 9 AM
Recommendation

Keep an inventory of any scheduled jobs or webhooks created through this skill, document owners, and include rollback or disable steps.