platform-engineer
ReviewAudited by ClawScan on May 4, 2026.
Overview
This is an instruction-only platform engineering advisor with no code or required credentials, but its guidance can affect infrastructure and access controls if connected to powerful tools.
This skill appears coherent and instruction-only. It is reasonable to install for platform-engineering assistance, but do not let it directly modify production infrastructure, access controls, GitOps repositories, or multi-cluster configurations without normal human review, scoped permissions, and rollback plans.
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.
If used with real infrastructure tools, the agent could help create or change environments, services, databases, scaling settings, or access workflows.
These are broad operational actions that can mutate infrastructure or account access if the agent is connected to deployment or platform tools. They are expected for the stated platform-engineer purpose, but users should notice the potential impact.
Self-service capabilities: - Environment provisioning - Database creation - Service deployment - Access management - Resource scaling
Use least-privilege tool access and require human approval for production deployments, access changes, resource scaling, or other irreversible platform changes.
Poorly scoped use could result in inappropriate access changes or weakened separation between teams or tenants.
The skill explicitly covers RBAC, isolation, and audit-related platform functions. No credentials are requested by the artifact, but these topics involve privileged administrative boundaries when acted on.
Platform architecture: - Multi-tenant platform design - Resource isolation strategies - RBAC implementation - Cost allocation tracking - Usage metrics collection - Compliance automation - Audit trail maintenance
Only grant the agent access to the specific platform, repository, or environment being worked on, and have a human administrator review RBAC or access-control changes.
A mistaken platform or GitOps change could affect multiple services, teams, or clusters.
GitOps and multi-cluster synchronization can propagate a single bad configuration change across shared infrastructure. The inclusion is purpose-aligned, but the blast radius can be large.
GitOps implementation: - Repository structure design - Branch strategy definition - PR automation workflows - Approval process setup - Rollback procedures - Drift detection - Secret management - Multi-cluster synchronization
Use staged rollouts, pull-request review, automated validation, and documented rollback plans before applying generated platform changes.
A user might trust unverified claims about completion, adoption, provisioning speed, or developer satisfaction.
The artifact includes a canned completion message with specific success metrics. If repeated without verification, it could overstate what was actually delivered.
Delivery notification: "Platform engineering completed. Delivered comprehensive internal developer platform with 95% self-service coverage, reducing environment provisioning from 2 weeks to 3 minutes..."
Treat this as a reporting template only; ask the agent to cite actual changes, measurements, pull requests, dashboards, or test results before accepting success claims.