The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes potentially untrusted external data such as exception messages, stack traces, and log entries via tools like posthog:query-error-tracking-issue-events and posthog:query-logs. This creates an indirect prompt injection surface where malicious instructions embedded in log data could attempt to influence the agent's synthesis or subsequent behavior.
Ingestion points: Data retrieved from error tracking and logging APIs (SKILL.md).
Boundary markers: The instructions do not define explicit delimiters or instructions to ignore embedded commands within the ingested content during synthesis.
Capability inventory: Tools available for execution include posthog:execute-sql and posthog:query-session-recordings-list (SKILL.md).
Sanitization: No explicit sanitization or filtering of retrieved log content is defined.
[COMMAND_EXECUTION]: The skill implements dynamic command generation by constructing SQL queries for the posthog:execute-sql tool through string interpolation of identifiers like issue_id and session_id obtained from previous tool outputs (SKILL.md).
[NO_CODE]: The skill is composed entirely of markdown instructions and does not include any standalone executable script files.

investigating-error-issue