The Agent Skills Directory

[SAFE]: The skill consists of a collection of best practices and security guidelines for Python development within the LlamaFarm project.
[SAFE]: The documentation explicitly warns against dangerous coding patterns such as the use of eval(), exec(), shell=True in subprocesses, SQL injection, path traversal, and unsafe deserialization (e.g., pickle.load() or broad yaml.load()).
[SAFE]: The skill provides search patterns (using grep/rg) for developers or agents to identify security vulnerabilities and non-idiomatic code in the codebase.
[SAFE]: No malicious code execution, data exfiltration, or obfuscation techniques were identified in the provided markdown files.
[SAFE]: The configuration in SKILL.md appropriately limits the agent's capabilities to reading and searching files (Read, Grep, Glob), which is consistent with its role as a code review and best practices guide.

python-skills