The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from marketing briefs and product offers, which could contain hidden malicious instructions designed to manipulate the agent's behavior.
Ingestion points: The skill ingests marketing briefs, product offers, and data from icp-deep-scanner (SKILL.md).
Boundary markers: The instructions lack explicit delimiters or warnings to ignore embedded instructions within the ingested data.
Capability inventory: The agent has access to Bash, Write, and WebFetch tools, which could be targeted by a successful injection attack.
Sanitization: There is no mention of input validation, filtering, or sanitization of the external content.
[COMMAND_EXECUTION]: The skill utilizes the Bash tool to run shell commands, specifically for interacting with the hyperframes-cli toolset.
Evidence: Use of npx hyperframes lint and npx hyperframes preview in SKILL.md.
[EXTERNAL_DOWNLOADS]: The skill relies on external software packages that are retrieved and executed at runtime.
Evidence: Mentions of using npx hyperframes and edge-tts for video rendering and neural voice generation (SKILL.md).

hyperframes-ad-director