The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits a vulnerability surface for indirect prompt injection by ingesting and processing untrusted data from external sources.
Ingestion points: External data is retrieved from user-provided arguments and GitHub issue bodies (via the gh CLI tool).
Boundary markers: No explicit delimiters or boundary markers (e.g., XML tags or "ignore embedded instructions" headers) are used when interpolating this untrusted data into task descriptions or subagent prompts.
Capability inventory: The skill has access to the Bash tool for shell execution, the Write tool for file modification, and the ability to spawn subagents.
Sanitization: No sanitization, escaping, or validation logic was found for the data ingested from external sources before it is used in prompt construction.
[COMMAND_EXECUTION]: The skill executes local shell scripts to perform diagnostic and analysis tasks on the repository.
Evidence: It invokes scripts/detect-plan-context.sh and scripts/analyze-impact.sh using the Bash tool. These scripts perform routine operations such as checking git diffs, branch names, and commit counts.
Evidence: It utilizes the GitHub CLI (gh) to retrieve issue metadata when an issue reference is provided.

visualize-plan