The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local command-line tool obsidian-wiki memory sync IMPORT in Step 5 to update vault metadata and synchronize internal state after an import operation.
[INDIRECT_PROMPT_INJECTION]: The skill serves as a vector for indirect prompt injection by ingesting external, potentially untrusted data and incorporating it into the vault's knowledge base.
Ingestion points: Ingests data from graph.json files (nodes, summaries) and OKF bundle markdown files (full page bodies) as described in Step 1 and Step 4.
Boundary markers: The instructions do not define delimiters or specific "ignore" warnings to prevent the agent from obeying instructions embedded within the imported summaries or page bodies.
Capability inventory: The skill has the capability to write and modify files within the user's vault and execute the obsidian-wiki CLI tool.
Sanitization: There is no evidence of sanitization or filtering of the natural language content (e.g., the summary or body text) before it is written into the Markdown files of the vault.

wiki-import