The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands (git rev-parse, git fetch, git pull) to synchronize the local project repository with its remote origin before applying any code fixes in Phase 4. While this is a standard developer workflow, it involves the execution of shell commands and interaction with the network.
[INDIRECT_PROMPT_INJECTION]: The skill ingests large amounts of untrusted data from the audited project's source code and metadata, creating a surface for indirect prompt injection. Maliciously crafted content in source files or app descriptions could attempt to influence the auditor's verdicts or the fixer agent's operations. Ingestion points: Project Explorer reads all project source code, Info.plist files, and metadata (e.g., description.txt). Boundary markers: No explicit instructions are provided to the agent to treat ingested file content as non-instructional or to use delimiters to prevent injection. Capability inventory: The Fixer agent has the capability to modify source files and Info.plist; the Report Writer writes markdown files to the project directory. Sanitization: No specific sanitization or filtering of external data before interpolation into prompts is documented.

appstore-review-checker