The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests external data from GitHub issue bodies via the gh issue view command. This data is used to extract acceptance criteria and drive subsequent agent actions.
Ingestion points: Issue body extraction in SKILL.md (Step 1).
Boundary markers: None identified; external content is processed directly as instructions for test planning.
Capability inventory: The skill has access to the Bash tool and can execute pnpm, playwright, and gh commands.
Sanitization: The skill does not specify any sanitization for the content extracted from issue bodies.
[COMMAND_EXECUTION]: The skill instructs the agent to run shell commands like pnpm test --grep "[test pattern]" using strings parsed from the issue body. A maliciously crafted issue could attempt to inject shell commands through these parameters.
[DYNAMIC_EXECUTION]: The skill triggers the execution of local project code through test runners (pnpm, playwright). This execution is dynamic and depends on the content of the repository being tested.
[DATA_EXFILTRATION]: Automated verification reports post test outputs to GitHub issue comments. This creates a risk of leaking sensitive local information, such as environment details or file snippets included in test failures, to an external platform.

acceptance-criteria-verification