The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes project-specific data to generate dashboards and recommendations, which introduces a surface for indirect prompt injection if the project files are compromised.
Ingestion points: The skill reads .planning/STATE.md, .planning/ROADMAP.md, and content from phase directories using the Read, Glob, and Grep tools.
Boundary markers: The instructions do not specify any boundary markers or delimiters to separate untrusted file content from system instructions.
Capability inventory: The skill is permitted to use Bash for command execution and the Skill and Agent tools for dispatching further tasks, which could be exploited if the agent is misled by malicious data in project files.
Sanitization: No explicit sanitization or validation of the ingested file content is mentioned.

gsd-manager