The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill maintains context by reading user-controlled files within the workspace, which represents a potential attack surface for indirect prompt injection.\n
Ingestion points: Reads MISSION.md, RESOURCES.md, learning-records/*.md, and NOTES.md (as described in SKILL.md) to ground its teaching behavior.\n
Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within these user-supplied files.\n
Capability inventory: The skill is capable of writing files to the local directory (lessons, assets, glossary) and performing network operations via the vg deploy command.\n
Sanitization: There is no evidence of sanitization or filtering of the content read from these markdown files before they are processed by the agent.\n- [EXTERNAL_DOWNLOADS]: The skill references and integrates with external game development tools and libraries.\n
The instructions reference standard engines such as Phaser and Three.js for game logic.\n
It acknowledges adaptation from the mattpocock/skills GitHub repository.\n- [COMMAND_EXECUTION]: The skill utilizes specific CLI tools for project workflow.\n
Mentions the use of vg generate for asset creation and vg deploy for project deployment to the vibedgames platform. These are external command-line operations integral to the skill's primary teaching purpose.

teach-me