The Agent Skills Directory

[SAFE]: The skill operates as a career advisor and does not contain any malicious instructions or dangerous code execution patterns. It uses standard file operations and web search tools for their intended purpose.
[PROMPT_INJECTION]: The skill incorporates an indirect prompt injection surface due to the ingestion of untrusted external data (Category 8).
Ingestion points: External data is ingested via the WebSearch tool to gather market benchmarks and success rates in the non-linear-careers.md and ai-readiness.md files. It also reads a local file career-helper-preferences.md for accessibility settings.
Boundary markers: The instructions do not define specific delimiters (e.g., XML tags or backticks) to isolate the data retrieved from the web from the agent's instructions.
Capability inventory: The skill's capabilities are limited to WebSearch, Glob, FileRead, and FileWrite. It lacks access to high-risk tools such as shell execution or external network requests (other than search).
Sanitization: The skill limits the influence of the local preferences file by specifying that only specific boolean flags from the YAML frontmatter should be applied.

career-transitions