The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the web through WebSearch queries to generate project verdicts and GitHub issue content. This creates a vulnerability surface where malicious instructions embedded in external web content could influence the agent's behavior.
Ingestion points: Web search results (Step W02) and external references.
Boundary markers: Absent. The skill does not provide delimiters or instructions for the agent to disregard potential instructions within the retrieved web data.
Capability inventory: GitHub CLI operations (gh issue create, gh issue comment) and file system writes to project reports.
Sanitization: Absent. There is no requirement for the agent to sanitize or validate the external content before incorporating it into the issue tracker or HTML reports.
[COMMAND_EXECUTION]: The skill uses the GitHub CLI (gh) to perform repository management tasks, such as creating issues and adding comments. Specifically, it executes commands like gh issue create --assignee Rheinmir and gh issue comment. These operations rely on the existing authentication of the gh tool in the environment.
[EXTERNAL_DOWNLOADS]: The skill performs multiple web searches to retrieve updates on agent frameworks, memory engineering, and security from external sources to compile its reports.

frontier-scan