The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified in the skill's instructions or logic.
[DATA_EXPOSURE]: The skill performs read and write operations on local project files located in the docs/ directory (e.g., docs/world-lore.md, docs/quest-registry.md). These operations are consistent with the skill's stated purpose of managing game design documentation and do not involve sensitive system files, environment variables, or credentials.
[PROMPT_INJECTION]: No attempts to bypass safety filters or override agent behavior were found. The use of "Hard Constraints" is task-oriented, ensuring the agent follows the designated game design workflow and phases.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external files (docs/*.md) which could technically serve as an injection surface if the files are attacker-controlled. However, the skill treats this data as context for creative writing and registry management without executing code or performing sensitive network operations based on the content.
Ingestion points: Reads content from docs/world-lore.md, docs/quest-registry.md, and docs/mvp-first-draft.md in Phase 0 and Phase 1.
Boundary markers: Absent.
Capability inventory: File reading and appending (docs/quest-registry.md). No network access or subprocess execution.
Sanitization: Absent.
[REMOTE_CODE_EXECUTION]: The skill does not download external scripts, execute shell commands, or use dynamic code execution functions like eval.

game-quest