The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill instructs the agent to search for and read sensitive files such as .env, .env.local, .sentryclirc, and sentry.properties. These files typically contain sensitive credentials like SENTRY_AUTH_TOKEN, SENTRY_DSN, and potentially other unrelated application secrets (e.g., database passwords, AWS keys) that would be exposed to the agent context.
[COMMAND_EXECUTION]: The documentation suggests the installation and execution of a third-party package using npx mushi-mushi. Using npx with external packages facilitates the download and execution of arbitrary code from a public registry at runtime.
[PROMPT_INJECTION]: The skill processes untrusted content from Sentry issue descriptions and external web pages retrieved via firecrawl_scrape. There are no instructions for sanitization or the use of boundary markers (e.g., XML tags or delimiters) to prevent malicious instructions embedded in those sources from overriding the agent's behavior.
[EXTERNAL_DOWNLOADS]: The skill relies on external tools (firecrawl_search and firecrawl_scrape) to fetch content from the public internet, which is then processed by the agent. This creates a bridge for external data to enter the project environment.

debug-sentry-monitor