The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill instructs the agent to fetch documentation summaries from https://context7.com/silentvoid13/templater/llms.txt. This domain is not an official source for the Templater plugin and the content is not hosted on a trusted or well-known service, posing a risk of unverifiable content ingestion.
[INDIRECT_PROMPT_INJECTION]: The skill facilitates the reading of existing vault notes and metadata to generate or debug templates, creating a vulnerability to malicious instructions embedded in the processed files.
Ingestion points: The skill uses obsidian read file (via external tool integration) and references tp.file.content to ingest note data into the agent's context.
Boundary markers: Absent. The instructions do not provide specific delimiters or warnings to ignore instructions found within the processed notes.
Capability inventory: The skill allows the agent to generate new files and modify existing ones through tp.file.create_new, tp.file.rename, and tp.file.move, which could be exploited if malicious instructions are followed.
Sanitization: Absent. There are no mechanisms described to sanitize or validate the content retrieved from the vault before it is used to generate executable Templater snippets.

obsidian-templater