The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill reads and processes content from multiple external files which could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: Processes content from phase plans (PLAN.md), summaries (SUMMARY.md), roadmaps (ROADMAP.md), and all source code files in the src/ directory.
Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the content of these files as untrusted data or to ignore embedded instructions.
Capability inventory: The skill executes shell commands (PowerShell) to search for patterns and check file existence. It also has the capability to write a VERIFICATION.md file based on the parsed data.
Sanitization: There is no evidence of sanitization, filtering, or escaping of the content read from files before it is processed by the agent to determine the verification outcome.
[COMMAND_EXECUTION]: The skill utilizes PowerShell commands to perform its core functions of file system inspection and content analysis.
Evidence: The instructions include automated steps using Get-ChildItem, Select-String, Test-Path, and Get-Content. While these are used for inspection purposes, they represent a significant capability that could be misused if the agent's environment is not properly restricted.

verifier