CitrineOS Assistant
Analysis
The skill is mostly coherent for CitrineOS administration, but it gives the agent direct installation and operational-control instructions, including starting transactions or resetting charging stations, without explicit confirmation safeguards.
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.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
Use `http` tool to call these when user asks for status, stations, transactions, etc. ... `Start transaction` → POST Message API `RequestStartTransaction` ... `Reset station` → POST Message API `Reset`
The skill instructs the agent to use HTTP calls for operational EV charging actions, including starting transactions and resetting stations, without an explicit confirmation or scoping requirement before making changes.
Clone: `git clone https://github.com/citrineos/citrineos-core` ... `npm run install-all && npm run build` ... `docker-compose -f docker-compose.yml up -d`
The skill guides the agent through local build and Docker startup commands. This is expected for a CitrineOS installation assistant, but it still runs project scripts and starts containers on the user's system.
