The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python script build_indexes.py to automate the indexing of notes. This script is located within the shared project structure and is used for internal vault maintenance.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it processes user-provided research data and synthesis packs to generate drafts. However, the operations are limited to local vault management, and no active exploitation patterns were found.
Ingestion points: Synthesis Packs and Daily Notes accessed via Stage 1 and Stage 2 workflows.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands in the source research.
Capability inventory: The agent can perform file system writes to the 01_Daily_Notes, 02_Topic_Notes, and 00_Index directories, and execute the local build_indexes.py script.
Sanitization: No sanitization or filtering of research content is specified before processing.

loom-write