The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/openocd_flasher.py executes the openocd command-line utility via subprocess.run. While it uses list-based arguments to mitigate shell injection, the script constructs these commands using inputs derived from the local environment and user arguments.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by automatically scanning the workspace for openocd*.cfg files and parsing .vscode/launch.json to extract flashing parameters. A malicious repository could contain crafted configuration files designed to inject unexpected arguments into the OpenOCD process.
Ingestion points: The scan_openocd_configs function in scripts/openocd_flasher.py reads configuration paths and server arguments from .vscode/launch.json and the file system.
Boundary markers: No delimiters or instructions are used to separate untrusted configuration data from the command logic.
Capability inventory: The skill possesses the capability to execute system commands and perform file system discovery through scripts/openocd_flasher.py.
Sanitization: The script does not validate or sanitize the strings retrieved from the workspace configuration files before appending them to the command-line argument list.

flash-openocd