The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on the execution of multiple local Python scripts (e.g., scripts/moz_api.py, scripts/commoncrawl_graph.py, scripts/verify_backlinks.py) and a shell script (./extensions/dataforseo/install.sh) to interface with APIs and manage the local environment.
[EXTERNAL_DOWNLOADS]: The skill references an installation script (./extensions/dataforseo/install.sh) for an external extension, which may download additional components to the system.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection through the ingestion of untrusted external data.
Ingestion points: Data is retrieved from the Moz, Bing, and Common Crawl APIs. Additionally, the verify_backlinks.py tool crawls external URLs to verify link presence and anchor text.
Boundary markers: There are no defined delimiters or 'ignore instructions' for the data retrieved from external web sources or APIs, allowing potentially malicious content to influence the agent's behavior.
Capability inventory: The skill can execute shell commands through its various scripts and perform file write operations to the .seo-cache/ directory.
Sanitization: The 'Pre-Delivery Review' section provides a checklist to ensure data accuracy and consistency (e.g., checking for reciprocal links or semantic headings), but it lacks specific security filters to sanitize input against prompt injection attacks.

seo-backlinks