The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from external sources.
Ingestion points: The url parameter used in the /api/transcribe, /api/analyze-image, and /api/convert endpoints (found in SKILL.md) allows the agent to fetch content from any platform.
Boundary markers: The skill does not define specific delimiters or instructions to the agent to disregard potentially malicious instructions embedded in the resulting transcriptions or OCR text.
Capability inventory: The skill facilitates network requests to the proxy.obul.ai domain but does not include scripts for local file system access or shell command execution.
Sanitization: There is no evidence of filtering or sanitizing the data retrieved from external URLs before it is presented to the agent.
[DATA_EXFILTRATION]: The skill transmits the user's OBUL_API_KEY and requested media URLs to the vendor-owned domain proxy.obul.ai. This is the documented and expected method for using the service.
[NO_CODE]: This skill contains no executable code, scripts, or binary files. It consists solely of markdown documentation and API request templates for the cnvrt.ing service.

obul-cnvrting