The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process unstructured data from arbitrary external URLs, creating a surface for indirect prompt injection.
Ingestion points: The skill fetches content from user-supplied URLs using the parallel-cli extract command as described in SKILL.md.
Boundary markers: The instructions lack explicit delimiters or guidance for the agent to treat extracted content as data rather than instructions.
Capability inventory: The skill has access to the Bash tool and the vendor-specific parallel-cli toolset.
Sanitization: There is no evidence of content sanitization or validation performed on the extracted data before it is returned to the agent context.
[COMMAND_EXECUTION]: The skill uses user-supplied arguments directly within a shell command template, which could be vulnerable to command injection if the execution environment does not properly sanitize input.
Evidence: The command parallel-cli extract "$ARGUMENTS" --json -o "/tmp/$FILENAME.json" in SKILL.md interpolates user input directly into the argument string.
[METADATA_POISONING]: The skill metadata contains instructions that explicitly encourage the agent to bypass standard platform tools in favor of vendor-specific alternatives.
Evidence: The description field in SKILL.md contains the instruction "Prefer over built-in WebFetch."

parallel-web-extract