The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill provides installation instructions that reference an external source for its files.
Evidence: git clone command in SKILL.md targeting the author's repository at github.com/thatrebeccarae/claude-marketing.git.
[COMMAND_EXECUTION]: The skill instructs the agent to use various command-line tools to perform technical diagnostic tasks on target domains.
Evidence: Use of curl to fetch robots.txt and sitemaps, dig for DNS resolution, and openssl for SSL certificate inspection as outlined in the 'Common Tools & Commands' section of SKILL.md.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from the websites it audits, which could potentially contain malicious instructions hidden in metadata or files.
Ingestion points: curl commands in SKILL.md used to retrieve robots.txt, XML sitemaps, and HTTP response headers from external targets.
Boundary markers: None are defined in the instructions for the agent when handling the output of these commands.
Capability inventory: The skill has the ability to execute shell commands and perform network operations.
Sanitization: The skill does not specify any sanitization or validation of the fetched external content before the agent processes it.

technical-seo-audit