The Agent Skills Directory

[PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks because it retrieves and acts upon external data from GitHub PR comments.
Ingestion points: Comments are fetched via get-pr-review-comments and get-pr-discussion-comments as documented in references/commands.md.
Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from following instructions that might be embedded within the retrieved PR comments.
Capability inventory: The skill allows the agent to execute shell commands via the erk utility, such as resolve-review-thread and post-pr-inline-comment.
Sanitization: No sanitization or filtering of the fetched comment content is specified before the agent evaluates it for classification and response generation.
[COMMAND_EXECUTION]: The skill uses the erk CLI for pull request management. This tool is treated as a legitimate vendor resource provided by the skill author for its intended purpose.

pr-operations