The Agent Skills Directory

[Prompt Injection] (HIGH): Vulnerable to Indirect Prompt Injection. The skill is designed to ingest and process technical documentation (untrusted data) from the file system.
Ingestion points: Uses Read and Grep to access project files, which may contain attacker-controlled content.
Boundary markers: Absent. There are no instructions to the agent to treat external content as data rather than instructions.
Capability inventory: Grants the agent Bash, Write, and Edit tools.
Sanitization: Absent. No logic is provided to sanitize or validate content before it is processed or used to generate outputs.
[Command Execution] (HIGH): The skill requests access to the Bash shell. While no malicious scripts are included in the skill definition itself, the combination of text-processing capabilities and terminal access allows for the execution of arbitrary commands if an indirect injection attack is successful.

code-of-conduct-generator