The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to read and aggregate content from summary and verification files within the project's planning directory, creating a surface for indirect prompt injection.
Ingestion points: The skill uses glob patterns to read files matching .planning/phases/*/*-SUMMARY.md and .planning/phases/*/*-VERIFICATION.md as specified in SKILL.md.
Boundary markers: There are no explicit instructions or delimiters provided to the agent to treat the content of these files as data only or to ignore embedded instructions.
Capability inventory: The skill has access to powerful tools such as Bash, Agent, and Write, which could be misused if the agent obeys instructions found within the audited files.
Sanitization: No sanitization, filtering, or validation of the processed files is mentioned in the instructions.

gsd-audit-milestone