The Agent Skills Directory

[Indirect Prompt Injection] (HIGH): The skill creates a vulnerability by instructing the agent to ingest external data while possessing high-privilege capabilities.
Ingestion points: Data returned from mcp__context7__query-docs (external documentation/web content).
Boundary markers: Absent. There are no instructions for the agent to treat fetched documentation as data rather than instructions, nor are there delimiters specified for external content.
Capability inventory: The skill allows the use of Bash, Edit, and Write tools. This allows the agent to execute shell commands and modify the local filesystem.
Sanitization: Absent. The skill does not provide mechanisms to validate, filter, or sanitize the content retrieved via the documentation tools.
Risk: An attacker who controls a documentation source (or performs a man-in-the-middle/poisoning attack on the repository/website) could embed instructions within the text that the agent might obey, leading to unauthorized command execution or file modification.

fluent-assertions