The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting data from external sources and other skill outputs.
Ingestion points: Reads contract data from ~/.claude/plugins/config/claude-for-legal-zh/commercial-legal/renewal-register.yaml. It also retrieves data via handoffs from the saas-msa-review skill and performs scans of external Contract Management Systems (CLM) and electronic signature tools.
Boundary markers: The instructions do not specify the use of delimiters or "ignore" instructions when processing untrusted contract fields (e.g., counterparty names, notes, agreement titles).
Capability inventory: The skill has file system read/write access to its configuration/registry directory and can invoke external tools via MCP to fetch agreement data.
Sanitization: No sanitization, validation, or escaping of the external data is mentioned before it is processed or presented to the agent.

renewal-tracker