The Agent Skills Directory

[SAFE]: The skill provides standard development guidelines, examples of Haskell code, and GHC optimization flags. No malicious behaviors were found.
[DATA_EXFILTRATION]: No network operations, sensitive file access, or credential harvesting patterns were identified.
[PROMPT_INJECTION]: The instructions focus on Haskell coding standards and do not attempt to override agent safety constraints or extract system prompts.
[REMOTE_CODE_EXECUTION]: There are no external downloads, remote script executions, or dynamic code evaluation patterns.
[OBFUSCATION]: The content is clear and readable with no hidden characters, encoded payloads, or steganographic patterns detected.
[COMMAND_EXECUTION]: The skill mentions GHC flags like -O2 and -ddump-simpl in an educational context for local compilation; it does not instruct the agent to execute arbitrary or dangerous shell commands.

practical-haskell