The Agent Skills Directory

[PROMPT_INJECTION]: No evidence of malicious instructions or attempts to override the AI agent's system prompt was found. The skill uses instructional language to guide the AI in generating structured interview content, which is consistent with its stated purpose.
[DATA_EXPOSURE]: The skill does not access sensitive local files, environment variables, or hardcoded credentials. It operates within the context of the user-provided role and candidate information.
[REMOTE_CODE_EXECUTION]: No scripts, external package installations, or remote execution patterns were detected in the skill files. All provided files are static markdown or JSON metadata.
[COMMAND_EXECUTION]: The skill does not utilize any shell commands or subprocess execution patterns. It relies entirely on the AI's natural language generation capabilities.
[DATA_EXFILTRATION]: No network operations or external data transmission patterns were identified. The output remains within the user's session.
[INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it is designed to process untrusted external data in the form of 'Candidate materials (resume/portfolio)' (referenced in SKILL.md and references/INTAKE.md).
Ingestion points: Candidate resumes and portfolios provided by the user.
Boundary markers: Absent; the skill does not explicitly instruct the agent to use delimiters or ignore instructions within the resume text.
Capability inventory: None; the skill does not have access to tools that execute code, write to the filesystem, or access the network.
Sanitization: Absent; no explicit validation of the input resume content is performed.
Risk Assessment: The risk is negligible because the skill lacks the capabilities to perform dangerous actions even if an injection occurs in the input data.

conducting-interviews