The Agent Skills Directory

[SAFE]: No prompt injection or behavior override patterns were detected. The instructions focus on strategic frameworks and brainstorming templates.
[SAFE]: The skill does not access sensitive local file paths (such as .ssh or .aws) and contains no hardcoded credentials or network exfiltration patterns.
[SAFE]: No obfuscation techniques such as Base64 encoding, zero-width characters, or homoglyph attacks were found in the provided files.
[SAFE]: The skill does not perform any remote code execution, package installations, or unauthorized script downloads. References to external tools like Crossbeam and PartnerStack are documented as pointers to local integration guides.
[SAFE]: No privilege escalation or persistence mechanisms were detected. The skill does not attempt to modify system configurations or establish recurring access.
[SAFE]: The skill is not vulnerable to dynamic context injection; it contains no shell execution placeholders or silent pre-execution commands.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it processes user-provided context about products and partners. However, the risk is negligible as the skill lacks executable capabilities like subprocess execution, file-writing, or network operations that could be exploited by malicious data.

co-marketing