The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to utilize standard shell tools such as ls, grep, and date. These are used legitimately within the workflow to research existing agents and establish accurate temporal context for document generation.
[SAFE]: The included validation script scripts/validate_agent.py is a safe utility that performs static regex-based analysis of markdown files. It uses only standard Python libraries (sys, re, pathlib) and does not engage in network operations or dynamic code execution.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates the creation of agents that ingest external data (user requirements) and utilize capable tools (Bash, Write). While this defines a potential attack surface for the agents produced using this guide, the skill itself does not contain malicious injection logic and provides a structured framework for validation.
Ingestion points: SKILL.md (Step 1.2 Gather Requirements).
Boundary markers: Not explicitly instructed in the design templates.
Capability inventory: The recommended toolset includes Bash, Read, Write, Edit, and Grep.
Sanitization: Not specified in the instructional content.

agent-creator