The Agent Skills Directory

[PROMPT_INJECTION]: The skill instructions focus on logical reasoning and cognitive psychology. No attempts to bypass safety filters, override system prompts, or use jailbreak techniques were found.
[DATA_EXPOSURE_AND_EXFILTRATION]: No sensitive file paths (e.g., .ssh, .aws), hardcoded credentials, or unauthorized network exfiltration patterns were detected. All external URLs point to the official deciqai.com domain or reputable academic sources.
[OBFUSCATION]: The content is entirely human-readable markdown. No Base64 encoding, zero-width characters, homoglyphs, or hidden text patterns were found in the skill or its examples.
[UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill does not include any dependency manifests (package.json, requirements.txt) or commands to download and execute remote scripts.
[PRIVILEGE_ESCALATION]: No commands requiring elevated privileges (sudo, chmod) or attempts to modify system configurations were identified.
[INDIRECT_PROMPT_INJECTION]: While the skill involves processing user-provided claims and evidence, it does not utilize unsafe interpolation or tool-chaining that would create a vulnerability surface. The intended use case is manual structured thinking.

confirmation-bias