The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The included scripts/inspect-project.py script reads and processes the content of various project files (Kotlin, Gradle, and XML manifests) using regular expressions to identify BLE implementation patterns. If an attacker places malicious instructions inside these project files (e.g., within comments or string literals), and the agent is asked to review the script's output or the file contents themselves, it could potentially influence the agent's behavior.
Ingestion points: The script ingests data from local files using Path.read_text() in functions inspect_build, inspect_manifest, and inspect_source within scripts/inspect-project.py.
Boundary markers: There are no explicit delimiters or boundary markers used when reading these files; the script treats all file content as raw text for regex analysis.
Capability inventory: The script is limited to local file system read operations and printing results to the console. It does not possess network access, file write capabilities, or subprocess execution permissions.
Sanitization: No sanitization or escaping is performed on the ingested text before it is processed by the regex engine or output to the terminal.

android-ble-inspector