The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes memory files which may contain untrusted data from previous user interactions.
Ingestion points: The agent reads MEMORY.md and various topic files within the memory directory to perform stock-taking and consolidation (SKILL.md).
Boundary markers: The instructions do not define specific delimiters or instructions to ignore potential commands embedded within the memory files.
Capability inventory: The skill's operations are limited to file reads and writes within the scoped memory directory; no shell commands, network requests, or dynamic code execution are performed.
Sanitization: There is no evidence of content sanitization or validation performed on the text during the merging or updating process.

consolidate-memory