The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local shell commands including curl, jq, mkdir, cat, and echo. These are used to create and read configuration files in the .listenhub/ directory and to interact with the vendor's API for content extraction and polling tasks.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes content from arbitrary external URLs provided by users or other skills. This content is later displayed as a preview and can be passed to other skills for further processing, creating a surface for indirect prompt injection if the source material contains malicious instructions.
Ingestion points: Content is retrieved from external URLs via the https://api.marswave.ai/openapi/v1/content/extract endpoint as described in SKILL.md.
Boundary markers: The instructions do not specify the use of clear delimiters or warnings to prevent the agent from interpreting extracted content as instructions.
Capability inventory: The skill utilizes shell command execution (via curl and jq) and file system write operations to save extracted results to the current directory.
Sanitization: No explicit sanitization or filtering of the extracted text content is mentioned before it is presented to the user or agent.
[EXTERNAL_DOWNLOADS]: The skill performs network requests to api.marswave.ai to submit extraction tasks and retrieve parsed content. This is the primary intended functionality of the skill using the vendor's infrastructure.

content-parser