Pocketalert
Analysis
This instruction-only skill is mostly aligned with sending Pocket Alert notifications, but it also documents credential-backed account-management and deletion commands that deserve review before use.
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.
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.
pocketalert messages send -t "Alert" -m "System update" -d all ... pocketalert apps delete <tid> ... pocketalert devices delete <tid> ... pocketalert webhooks delete <tid>
The documented command set includes broadcasting to all devices and deleting account resources, which is broader and more destructive than simply sending user-directed alerts.
Download from https://info.pocketalert.app/cli.html and extract to /usr/local/bin/
The setup relies on an external CLI download and privileged local install path, but the artifact provides no checksum, package pin, or install spec.
Checks whether tool use, credentials, dependencies, identity, account access, or inter-agent boundaries are broader than the stated purpose.
pocketalert auth <your-api-key> ... Configuration is stored at `~/.pocketalert/config.json`.
The skill depends on a Pocket Alert API key and local credential/config storage. This is expected for the service, but users should recognize that the agent can act with that account authority once authenticated.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
pocketalert webhooks create --name "GitHub Push" --message "Push to %repository.name%: %head_commit.message%"
Webhook templates can transform incoming third-party payload fields into push notifications, so data boundaries depend on the webhook source and template contents.
