The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) as it is designed to ingest and process data from external sources.
Ingestion points: Reads project configuration files (e.g., package.json, go.mod) and fetches content from external documentation URLs defined in research-playbook.md.
Boundary markers: The skill does not implement explicit delimiters or instructions to ignore potential injections within the researched documentation.
Capability inventory: The skill uses ls, cat, and grep for discovery and utilizes an agent tool (claude) to generate and commit new skill bundles to the local filesystem.
Sanitization: No explicit sanitization or validation of the retrieved documentation content is observed before it is used to generate new skill files.
[COMMAND_EXECUTION]: Employs standard, non-destructive bash commands such as ls, cat, and grep to identify project frameworks, language versions, and existing Sentry configurations. These operations are limited to local project discovery.
[EXTERNAL_DOWNLOADS]: Recommends the use of the official Sentry CLI wizard (npx @sentry/wizard@latest) for automated project setup. It also directs the agent to perform research at docs.sentry.io and official GitHub repositories. These are recognized as well-known and trusted services associated with the skill's author (getsentry).
[DATA_EXFILTRATION]: No unauthorized data transmission was detected. The skill handles sensitive configuration via standard environment variables (e.g., SENTRY_AUTH_TOKEN) and encourages the use of interactive, official Sentry tools for authentication, which is standard practice for this vendor.

sentry-sdk-skill-creator