cracked-profile
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute local shell commands such as
echo,date +%F,git config user.name, andgh api userto resolve data paths, get the current date, and infer user profile information (name and GitHub handle). - [DATA_EXFILTRATION]: The agent is directed to scan personal user directories (
~/Downloads,~/Documents,~/Desktop) for files matching*resume*or*cv*. While this is a functional requirement for creating a career profile, it involves accessing sensitive personal folders. The skill mitigates risk by instructing the agent to confirm the file path with the user rather than assuming it. - [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it reads content from user-provided resumes and local Markdown files (
profile.md,journal.md) to populate the agent's context. - Ingestion points: Reads
profile.md,journal.md, and external resume files (PDF, Docx, MD, Tex). - Boundary markers: No specific delimiters or "ignore embedded instructions" headers are defined for the file content.
- Capability inventory: The agent can execute shell commands (
git,gh,date) and perform file read/write operations. - Sanitization: No specific sanitization or validation of the ingested file content is mentioned.
Audit Metadata