The Agent Skills Directory

[PROMPT_INJECTION]: No evidence of instructions intended to bypass safety filters or override core agent behavior. The instructions focus on adopting a specific communication style (tactical empathy) for negotiation scenarios.
[DATA_EXPOSURE]: No access to sensitive file paths, environment variables, or hardcoded credentials detected.
[EXTERNAL_DOWNLOADS]: No network requests, URL references, or external package requirements found in any of the skill files.
[COMMAND_EXECUTION]: No shell commands, subprocess spawning, or system-level operations are present.
[OBFUSCATION]: All content is in cleartext markdown. No Base64, zero-width characters, or encoded strings were detected.
[PERSISTENCE]: No attempts to modify system configuration, shell profiles, or scheduled tasks.
[INDIRECT_PROMPT_INJECTION]: While the skill is designed to process user-provided communication drafts, it lacks any high-risk capabilities (network, file system, execution) that could be exploited via malicious input data.
[DYNAMIC_EXECUTION]: No use of eval, exec, or runtime code generation patterns.

strategic-communication