The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and summarize data from external Axiom datasets, which constitutes an indirect prompt injection surface. If a dataset contains malicious instructions within indexed fields (e.g., error messages, status codes), these could potentially influence the agent's behavior when generating the anomaly report.
Ingestion points: Data returned from axiom query operations described in SKILL.md (e.g., volume anomalies, new values, rare events).
Boundary markers: Absent. The skill does not instruct the agent to use specific delimiters or instructions to ignore potential commands embedded in the dataset content.
Capability inventory: Access to Bash (axiom CLI) for querying and configuration as defined in the allowed-tools frontmatter.
Sanitization: Absent. The skill provides direct APL queries for data retrieval without specifying sanitization or escaping of the resulting values before inclusion in the final report.

detect-anomalies