The Agent Skills Directory

[SAFE]: The skill uses standard package management and version control commands (npm, git) to facilitate the upgrade of the @linear/sdk. The allowed-tools configuration is appropriately scoped to these tasks using restricted bash commands.
[SAFE]: All external URLs point to legitimate resources, including the official Linear documentation (linear.app), the npm registry, and the author's verified personal and company domains.
[INDIRECT_PROMPT_INJECTION]: The provided audit script example demonstrates a pattern where local source code is ingested for analysis. This is a common pattern for coding assistance skills and is implemented safely here.
Ingestion points: The scripts/audit-linear-usage.ts example reads file content from the src/ directory.
Boundary markers: None; the script performs static regex matching on the file content for reporting purposes.
Capability inventory: The skill utilizes Read, Write, Edit, Bash, and Grep tools.
Sanitization: Content is matched against specific SDK-related patterns and logged to the console; it is not executed or passed back into a command generation loop.

linear-upgrade-migration