The Agent Skills Directory

[SAFE]: The skill consists entirely of Markdown documentation (SKILL.md, references/private-offer-desk.md) and evaluation data (evals/evals.json). It does not contain any executable scripts, binary files, or configuration that would trigger automated actions.
[SAFE]: No prompt injection patterns were detected. The instructions are structured as a professional playbook for business strategy and do not attempt to bypass safety filters, extract system prompts, or override agent behavior.
[SAFE]: There are no instances of data exposure or exfiltration. The skill does not access sensitive file paths (e.g., .ssh, .aws, .env), nor does it include hardcoded credentials or unauthorized network operations.
[SAFE]: No obfuscation techniques were identified. The content is written in clear, plain-text Markdown and JSON with no use of Base64 encoding, zero-width characters, or homoglyph substitutions.
[SAFE]: The skill does not perform any remote code execution or package installation. It does not utilize curl, wget, npm, or pip to fetch external code.
[SAFE]: No privilege escalation or persistence mechanisms are present. There are no commands involving sudo, system service modifications, or shell profile edits.
[SAFE]: While the skill involves an 'Interview' process that ingests untrusted user data, the skill lacks any capabilities (such as shell access, file writing, or network requests) that could be exploited via indirect prompt injection.

aws-marketplace-playbook