The Agent Skills Directory

[SAFE]: The skill consists of instructional markdown files and does not contain any executable code, scripts, or remote dependencies. No malicious patterns such as obfuscation, credential harvesting, or command injection were found.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data such as PRDs, design documents, and RFCs to generate task artifacts.
Ingestion points: External source documents (PRDs, RFCs, design docs) and repository code are processed by the agent in the 'Cut' and 'Ground' phases.
Boundary markers: The skill includes strict logical boundaries, such as the 'Refuse rather than guess' rule and the requirement for observable outcomes with concrete values.
Capability inventory: The skill instructs the agent to create task files in the '.tasks/' directory and read repository content. No direct shell execution or network exfiltration capabilities are utilized by the skill instructions themselves.
Sanitization: There are no explicit instructions for sanitizing or escaping content from external documents before inclusion in the generated task files.

tlc-plan