The Agent Skills Directory

[SAFE]: The skill is well-documented and its behavior matches its stated purpose. It orchestrates a workflow for resume customization without exhibiting malicious traits.
[COMMAND_EXECUTION]: The skill utilizes a bundled shell script (scripts/render.sh) to perform document conversions. The script is securely written, using proper variable quoting and standard system utilities (such as Pandoc and Chrome/Chromium) to generate output. No unauthorized or dangerous command execution was found.
[EXTERNAL_DOWNLOADS]: The skill provides the ability to fetch job descriptions from user-provided URLs. This is a legitimate functional requirement for JD analysis and does not involve the retrieval or execution of remote scripts or packages.
[DATA_EXFILTRATION]: Data access is limited to the user's career profile and job descriptions. Output files are stored locally, and versioning is handled through local git commits. There is no evidence of sensitive data being transmitted to external servers.
[PROMPT_INJECTION]: The skill processes external data (job descriptions), which represents a standard ingestion surface. The instructions focus on structured keyword extraction and achievement mapping, with no signs of malicious injection or safety guideline bypasses.

resume-tailor