The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/git_bisect_runner.py executes arbitrary shell commands provided by the user via the --test parameter. This is implemented in the test_commit function using subprocess.run(["bash", "-c", test_cmd]). While this is the intended primary purpose of the skill, it creates a high-privilege execution environment where any command the agent has permissions for can be run on the host system.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it processes and executes commands that may originate from untrusted external sources.
Ingestion points: The --test, --repo, --good, and --bad arguments in scripts/git_bisect_runner.py act as entry points for data that could be influenced by an attacker.
Boundary markers: None. The skill does not use delimiters or instructions to ignore embedded commands within the test string.
Capability inventory: The script has the capability to execute any shell command via bash -c and perform file system operations through git commands.
Sanitization: No sanitization or validation is performed on the test_cmd string before it is passed to the shell for execution.

git-bisect-assistant