The Agent Skills Directory

[COMMAND_EXECUTION]: The instructions in references/distillation-process.md direct the agent to execute shell commands such as ls and grep to explore the vault's namespace and search for existing notes. It also uses mv to retire source files to a _distilled/ directory and executes a local Python script scripts/check_distillation.py to validate the resulting output. Furthermore, it references an external local dependency generate_knowledge_index.py from a separate vault-maintenance skill.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and organize external data, such as transcripts and captures, which creates a potential attack surface for indirect prompt injection.
Ingestion points: The primary input for the skill is a capture, transcript, or draft defined in the argumentHint of SKILL.md.
Boundary markers: Absent. The skill does not instruct the agent to use specific delimiters or protective framing to distinguish between the source content and the system's operational instructions.
Capability inventory: The agent is granted capabilities to modify the file system (mv) and execute shell commands and local scripts.
Sanitization: Absent. There are no guidelines for sanitizing or filtering instructions that might be embedded in the source material before it is processed by the agent.

note-distillation