The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from meeting transcripts and vault entries to extract actions and update task databases.
Ingestion points: Reads meeting transcripts stored in the vault and project context files in the .context/ directory.
Boundary markers: The instructions do not specify the use of delimiters or protective headers when processing transcribed text to prevent the agent from following embedded instructions.
Capability inventory: The agent has access to Write and Edit tools to modify local files, and Bash(taskflow-cli *) to execute database queries and updates.
Sanitization: No content filtering or validation is performed on the ingested text before it is used to influence task creation or focus updates.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute taskflow-cli commands for vault operations. This is a primary function of the skill and is scoped to the specific CLI tool in the configuration.

task-management