The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests data from external sources, creating a potential attack surface for indirect prompt injection.
Ingestion points: The agent reads the current git branch name and fetches issue and team metadata from the Linear API via mcp__linear-server__get_issue and mcp__linear-server__list_issue_statuses.
Boundary markers: There are no explicit instructions to wrap external data in delimiters (e.g., XML tags) to prevent the agent from interpreting data as instructions.
Capability inventory: The skill's primary capabilities are updating issue states via mcp__linear-server__save_issue and reading git branch information.
Sanitization: The skill includes explicit instructions to escape regex metacharacters when processing branch names, which prevents malicious branch names from manipulating the issue-ID extraction logic.
[SAFE]: The skill adheres to the principle of least privilege by using a restricted Bash tool Bash(git:*) and standard MCP tools. All external operations are handled through authorized tools, and no arbitrary code execution or unauthorized data access patterns were detected.

linear-sync