The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external Sentry reports, which could potentially influence the agent's behavior if the data contains malicious instructions.
Ingestion points: Sentry API responses containing issue titles and event data are fetched via curl and processed in SKILL.md.
Boundary markers: No explicit delimiters are used to isolate the fetched data from the skill's instructions.
Capability inventory: The skill utilizes Bash, Write, and Edit tools to perform incident management tasks.
Sanitization: The skill extracts specific JSON fields using python3, providing a degree of structural integrity but lacking content-based filtering.
[COMMAND_EXECUTION]: Shell commands are employed for API interaction and data processing.
Evidence: SKILL.md includes various curl and python3 -c command sequences for communicating with Sentry.
[EXTERNAL_DOWNLOADS]: Fetches incident and performance data from a well-known service.
Evidence: The skill downloads information from Sentry's official API endpoints at sentry.io.

sentry-incident-runbook