The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill retrieves priorities (P0s) from external files (weekly reviews) to populate the daily plan. This creates a surface for indirect instructions to influence agent behavior.
Ingestion points: SKILL.md instructs the agent to pull data from the 'latest weekly review' note.
Boundary markers: The instructions lack delimiters or explicit directives to ignore potential commands embedded within the source notes.
Capability inventory: The agent possesses file system read/write access limited to the vault's daily directory as defined in the configuration.
Sanitization: There is no mention of sanitizing or validating the content extracted from the markdown files before processing.
[DATA_EXPOSURE]: The skill interacts with the local file system to manage markdown notes.
Path resolution is handled dynamically through configuration keys (e.g., layout.daily_dir), avoiding the exposure of hardcoded machine paths.
The ROLE.md file establishes a strict 'secretary' persona, explicitly forbidding the agent from editing external workspaces or implementing product code, which limits the scope of potential data exposure.

life-plan-day