The Agent Skills Directory

[PROMPT_INJECTION]: The skill features a surface for indirect prompt injection through data ingestion.
Ingestion points: The script scripts/validate-list.ts reads content from external JSON files provided as input.
Boundary markers: There are no instructions for the agent to use delimiters or ignore embedded instructions when reading list items.
Capability inventory: The agent has permissions to write files to the project directory and perform web-based research.
Sanitization: No sanitization or filtering logic is present to identify or neutralize instructions within the JSON data.
[COMMAND_EXECUTION]: The skill involves the execution of local Deno scripts to perform its primary functions.
Evidence: SKILL.md directs the agent to run deno run --allow-read scripts/validate-list.ts to analyze lists.
Context: The scripts are provided within the skill package and perform legitimate validation and merging operations as documented.

list-builder