The Agent Skills Directory

[DATA_EXPOSURE]: The skill reads local configuration files such as ~/.obsidian-wiki/config and .env to determine the location of the Obsidian vault. It subsequently accesses files within the vault, including .manifest.json, index.md, and individual markdown pages, to map knowledge provenance.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the local wiki vault to generate summaries and comparisons.
Ingestion points: Reads metadata and content from .manifest.json, index.md, and various wiki pages.
Boundary markers: None identified; the skill does not explicitly use delimiters to separate ingested data from agent instructions.
Capability inventory: Performs file read operations, local logging (file write), shell command execution via grep, and tool invocation of the impl-validator subagent.
Sanitization: No explicit sanitization or filtering of the ingested content is specified before it is processed or presented to the user.
[COMMAND_EXECUTION]: Utilizes the grep utility to search for specific topics within the vault's markdown files based on user input.

memory-bridge