The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data by scanning HTML, Vue, Svelte, and other template files within a project repository. If these files contain malicious hidden instructions, they could potentially influence the agent's behavior during the implementation of SEO fixes. Ingestion points include the scripts/audit_seo.py utility and manual review tasks assigned to the auditor agent. No explicit boundary markers or content sanitization steps are defined to mitigate this risk. The skill maintains significant capabilities, including the ability to write project files and execute shell commands.
[EXTERNAL_DOWNLOADS]: The skill depends on the website-agent-readiness skill, which is fetched from an external repository at github.com/luongnv89/skills. This dependency is managed via the asm tool and represents vendor-owned functionality as it shares the same author as the primary skill.
[COMMAND_EXECUTION]: The skill utilizes shell commands to manage repository state and perform audits. This includes the use of git fetch, git pull, and git stash for synchronization, as well as the execution of the bundled Python audit script.

seo-ai-optimizer