The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's primary functionality is built around executing shell commands via the postplus CLI tool. These commands include postplus research run, postplus research schema, postplus doctor, and postplus quote confirm to perform various Instagram research tasks, check status, and manage execution costs.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data from Instagram, including public comments, profile biographies, and post captions. This content is processed by the agent to identify themes, pain points, and audience voice, creating a surface where malicious instructions embedded in social media content could influence the agent's output.
Ingestion points: Content from result.json files generated by the CLI tools after fetching Instagram data (e.g., instagram-comments, instagram-profiles).
Boundary markers: The instructions do not specify explicit boundary markers or instructions for the agent to ignore potentially malicious content within the processed social media data.
Capability inventory: The skill can execute shell commands via the postplus CLI and read/write to the local filesystem for result persistence.
Sanitization: There is no evidence of specific sanitization or filtering logic applied to the external content before it is processed for analysis.

instagram-research