The Agent Skills Directory

[COMMAND_EXECUTION]: The skill interacts with the system using yy ledger and yy ledger artifact commands to record development requirements and manage task records. This activity is consistent with the skill's stated purpose for Kanban management and uses resources associated with the skill's author (yylo-dev).
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the local environment and user input, creating a potential vector for indirect prompt injection.
Ingestion points: The skill is instructed in Step 1 to read project instructions and relevant product code from the integration or feature worktree, as well as processing user-provided features, constraints, and criteria via $ARGUMENTS.
Boundary markers: Absent. There are no explicit instructions or delimiters used to separate the ingested data from the agent's core instructions or to ignore embedded commands within the read files.
Capability inventory: The skill utilizes shell directives to execute commands via the yy CLI tool, which could be misused if a malicious instruction is encountered in the ingested content.
Sanitization: Absent. The skill does not specify any validation, filtering, or escaping of the content read from the product code or instructions before it is processed.

plan-ledger-tasks-yylo