The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local CLI commands basic-memory and bm to retrieve version information and hook status. These tools are vendor-owned resources directly related to the skill's primary diagnostic function.
[EXTERNAL_DOWNLOADS]: The skill uses the uvx runner to verify the basic-memory package version. This involves communicating with the Python Package Index (PyPI), a well-known service, to fetch vendor-owned package metadata.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external notes that could potentially contain malicious instructions, creating an attack surface.
Ingestion points: Metadata (titles and permalinks) retrieved from the search_notes tool (SKILL.md).
Boundary markers: No explicit delimiters or instructions to ignore embedded content are provided in the presentation template.
Capability inventory: The skill has access to shell command execution (basic-memory, bm, uvx) and file reading (settings.json).
Sanitization: No specific sanitization, validation, or escaping of the note metadata is performed before it is rendered in the final report.

bm-status