The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill uses a local script scripts/resolve-source-root.mjs to determine the environment's file system structure and locate the canonical repository for mission tracking.
[DYNAMIC_EXECUTION]: The skill executes the xref command-line tool with various flags (--seeds, --repo, --prioritize, --cluster) to process GitHub repository metadata and generate rankings for potential issues.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources (GitHub issues and comments), which acts as a potential surface for indirect injection attacks.
Ingestion points: External GitHub issue titles, labels, comment counts, bodies, and user comments from a repository backlog.
Boundary markers: The instructions do not explicitly define boundary markers or delimiters to isolate untrusted content from agent commands.
Capability inventory: The skill is capable of writing state files to the local filesystem (foundry/missions/) and executing localized project tools (xref, resolver scripts).
Sanitization: There is no explicit mention of sanitizing or filtering instructions that might be embedded within the issue content retrieved from the backlog.

issue-intake