The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill provides instructions to install the Sentry CLI by downloading and executing a script directly from Sentry's official domain via a shell pipe.
Evidence: curl https://cli.sentry.dev/install -fsS | bash in SKILL.md.
[EXTERNAL_DOWNLOADS]: The skill references external installation sources for the Sentry CLI, including a dedicated install script and the npm registry.
Evidence: References to https://cli.sentry.dev/install and npm install -g sentry.
[COMMAND_EXECUTION]: The skill is designed to facilitate the execution of Sentry CLI commands, which interact with local and remote resources.
Evidence: Extensive list of sentry commands including auth, org, project, issue, and api.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data from Sentry issues and events. This data is untrusted and could contain malicious instructions meant to influence the AI agent's behavior.
Ingestion points: Data is ingested through commands like sentry issue list, sentry issue view, and sentry event view in SKILL.md.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to treat issue content as untrusted data.
Capability inventory: The skill has the ability to perform authenticated network requests via sentry api and execute various CLI commands.
Sanitization: There is no evidence of sanitization or filtering of the retrieved issue or event data before it is presented to the agent.

sentry-cli