The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions require the agent to execute several local shell scripts and binaries.
Evidence: Mentions and executes ./android_screencap.sh, ./build.sh, and ./screenshot.sh.
Evidence: Directly invokes adb logcat and grep commands to process device data.
[DYNAMIC_EXECUTION]: The skill performs runtime compilation of native code, which is a significant security risk as the source logic is not statically fixed.
Evidence: SKILL.md states that the app will "Build automatically if needed" and references main.swift and build.sh for compiling the native macOS app.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from an external device, which could be used to influence the agent's behavior.
Ingestion points: The agent processes output from adb logcat, specifically looking for lines prefixed with [APP].
Boundary markers: None. The skill does not implement delimiters or instructions for the agent to ignore potentially malicious content within the logs.
Capability inventory: The skill has the capability to execute shell scripts and interact with the filesystem (via screenshot.sh and adb).
Sanitization: There is no evidence of sanitization or validation of the log content before it is processed by the agent.

android-screen-capture