The Agent Skills Directory

[SAFE]: The skill operates entirely within a local environment. The provided Python script, scripts/claim_audit.py, uses only the standard library and performs string/regex-based analysis on user-provided text. It does not utilize any dangerous functions such as eval(), exec(), or subprocesses that interact with the network or sensitive files.
[SAFE]: The instructions in SKILL.md are focused on academic integrity and enforcing statistical constraints. There are no attempts to override agent safety guidelines or bypass system prompts. The allowed-tools configuration (Read and Bash) is appropriate for the skill's functionality of reading reference files and running the local linter.
[SAFE]: No obfuscation, data exfiltration patterns, or persistence mechanisms were detected across any of the files.

alterlab-ssci-inference-gate