Yes Tool
ReviewAudited by ClawScan on May 15, 2026.
Overview
This is a simple yes-like output tool, but it encourages blindly answering prompts and advertises safety limits that the included code does not actually implement.
Treat this as an unbounded 'yes' generator, not a safe automation controller. Before installing or using it, verify the implementation supports any limits you plan to use, and never pipe it into destructive, privileged, legal-consent, or account-changing commands without explicit human review.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
If used with the wrong command, it could approve actions the user did not review, including installs, deletions, or terms acceptance.
The documentation explicitly frames the tool as a way to auto-confirm prompts and includes a destructive command pipeline example, without requiring confirmation boundaries or warning about high-impact downstream actions.
Repeatedly output a string (default: "y") to automatically confirm prompts... Essential for unattended script execution... # Default usage... yes-tool | rm -rf dir/*
Only use this with explicit user approval and low-risk commands; avoid piping it into destructive, privileged, or terms-acceptance workflows unless the exact action and consequences are reviewed first.
A user or agent expecting a limited or rate-limited output could instead get an endless fast stream, causing hangs, flooding a pipeline, or continuing an unintended automation.
The only included implementation ignores all options and loops forever, contradicting SKILL.md's advertised controls such as '--count' and '--sleep'. Users may rely on safeguards that are not actually present.
s = sys.argv[1] if sys.argv[1:] else 'y' while True: print(s)
Do not rely on the documented options unless the implementation is fixed and tested; require a working count limit or external timeout before using it in automation.
