The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/create_report_packet.py executes git rev-parse --show-toplevel to programmatically determine the repository root. This is a standard and benign use of subprocess calls for repository management.
[COMMAND_EXECUTION]: The scripts/test_skill.py script uses subprocess.run to invoke the skill's internal scripts (create_report_packet.py and scan_site.py) as part of its validation and testing suite.
[EXTERNAL_DOWNLOADS]: The scripts/scan_site.py script makes a network request to https://isitagentready.com/api/scan to retrieve audit results. This targets the well-known service associated with the skill's primary function to retrieve scan data for a user-provided URL.
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it processes untrusted data.
Ingestion points: The skill reads local repository files (e.g., robots.txt, sitemap.xml) and fetches JSON data from the isitagentready.com API in scripts/scan_site.py.
Boundary markers: The instructions in SKILL.md and references/report-format.md define a structured reporting format using templates, which acts as a logical boundary, though no explicit security delimiters are used for the content of the files themselves.
Capability inventory: The skill has the capability to write files (creating the report packet), perform network operations (scripts/scan_site.py), and execute shell commands via git and local python scripts (scripts/create_report_packet.py, scripts/test_skill.py).
Sanitization: The skill performs minimal sanitization; it uses template replacement to insert variables into the report but does not explicitly filter or escape the content gathered from repository files or API responses.

isitagentready