The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it processes data from external files to influence agent output and file modifications.
Ingestion points: The skill reads world data and quest history from docs/world-lore.md, docs/quest-registry.md, and docs/mvp-first-draft.md. It also processes the quest name provided in the user's message.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when parsing content from the lore or registry files.
Capability inventory: The skill is permitted to read local files, create new files (docs/quest-registry.md), and append data to existing documentation. It does not have network access or general shell execution capabilities.
Sanitization: There is no explicit sanitization or validation of the content retrieved from external files before it is interpolated into the coherence report, quest brief, or registry updates.

game-quest