The Agent Skills Directory

[SAFE]: No security threats or malicious patterns were identified. The skill is limited to fetching and reporting cloud cost data via the specified Harness MCP server.- [INDIRECT_PROMPT_INJECTION]: The skill features a data ingestion surface through external cloud cost data and recommendations.
Ingestion points: Data is ingested via MCP tool calls fetching cost_breakdown, cost_recommendation, and cost_anomaly resources as described in SKILL.md.
Boundary markers: Absent; the instructions do not implement delimiters to isolate ingested cost data from the agent's internal instructions.
Capability inventory: The skill uses harness_execute for reporting feedback on anomalies in SKILL.md.
Sanitization: None detected; the skill relies on the structured nature of the Harness MCP responses.

analyze-costs