The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions provide bash commands for creating and removing marker files in ~/.claude/cooking/ using ps, mkdir, printf, and rm to track active execution.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core function of reading and synthesizing data from the auto-memory graph.
Ingestion points: Reads every .md file located under ~/.claude/projects/*/memory/ as described in SKILL.md.
Boundary markers: The skill lacks explicit instructions or markers to distinguish between data and potential instructions within the memory files.
Capability inventory: Performs broad file reads and writes a synthesized output to a file outside the memory directory.
Sanitization: No content sanitization or instruction-filtering is specified for the ingested markdown data.

dreaming-consolidator