The Agent Skills Directory

[PROMPT_INJECTION]: The skill demonstrates an indirect prompt injection surface by processing untrusted data from both user input and the codebase.
Ingestion points: The skill ingests untrusted data from user-provided arguments ($ARGUMENTS) and the local codebase via the Read, Grep, and Glob tools during the 'Explore and Diagnose' phase (file: SKILL.md).
Boundary markers: There are no explicit delimiters or instructions directing the agent to ignore or isolate instructions that might be embedded in the code files being analyzed.
Capability inventory: The skill possesses the capability to execute shell commands via the Bash tool and perform network-side-effect operations via the gh (GitHub CLI) tool.
Sanitization: No sanitization or escaping of the ingested content is performed before it is used to generate the GitHub issue body.
[COMMAND_EXECUTION]: The skill uses the Bash tool and GitHub CLI to perform autonomous repository actions. It explicitly instructs the agent to skip user review before filing issues ('MANDATORY: file the issue immediately ... do not ask the user to review a draft first'), which reduces human oversight for operations that modify the remote repository state.

triage-issue