The Agent Skills Directory

[SAFE]: The skill's primary function is to analyze performance data provided by the user in the form of JSON profile files. No malicious code, obfuscation, or unauthorized data access was detected.
[COMMAND_EXECUTION]: The skill utilizes standard command-line tools such as jq, grep, file, and gunzip for data processing and file identification. These operations are performed on user-provided local files and do not involve remote code execution or privilege escalation.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it processes untrusted data from profile files. However, the analysis is performed using structural tools like jq, which limits the risk.
Ingestion points: Processes user-provided .json, .reactprofile, and .cpuprofile files (specified in SKILL.md and rules/input-detection.md).
Boundary markers: No explicit instructions are present to disregard embedded instructions within the profiles, though the structural nature of the processing provides inherent protection.
Capability inventory: The skill uses jq, grep, file, and gunzip to process the files and invokes a separate confidence skill for result verification.
Sanitization: There is no explicit sanitization of the content within the profile files before processing.

profile-optimizer