Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
PUA Debugging
v1.0.0Forces exhaustive problem-solving using corporate PUA rhetoric and structured debugging methodology. MUST trigger when: (1) any task has failed 2+ times or y...
⭐ 0· 350·5 current·5 all-time
by@tanweai
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
Name/description promise exhaustive, proactivity-driven debugging; SKILL.md consistently instructs the agent to exhaustively search, read source, run validations and verify fixes. There is no mismatch between claimed purpose and required actions.
Instruction Scope
The instructions mandate using platform tools (search, file reads up to 50 lines of source, running tests/curl/commands, isolating environments) and to 'do tools-first, ask-later' before user prompts. While relevant to debugging, this grants the agent broad discretion to access files/execute commands and to probe environments without explicit per-task consent — a privacy and safety risk that should be constrained by runtime policies.
Install Mechanism
Instruction-only skill with no install spec and no code files — low file-system/install risk.
Credentials
Skill declares no env vars/credentials (appropriate). However, it expects access to platform tools (search, file I/O, command execution). Those expectations are plausible for debugging but should be limited to intended scopes and audited, since the skill does not explicitly restrict which files/paths or external services may be contacted.
Persistence & Privilege
always:false (good). The skill allows normal autonomous invocation; combined with its broad tool-use instructions this raises the blast radius if the agent is permitted to act without careful sandboxing or permission prompts.
What to consider before installing
This skill is coherent with a 'force persistence' debugging role but is potentially risky in practice. Before installing: (1) ensure your agent runtime enforces per-skill/tool permissions (limit which files/directories and which system commands it may run); (2) configure explicit consent rules so the agent asks before accessing sensitive paths or sending data externally; (3) test the skill in a sandboxed environment to observe behaviors and tune trigger conditions (it triggers on many user-stated frustrations and failure counts); (4) consider user-experience/ethical impact — the PUA-style rhetoric is aggressive and may be inappropriate for many contexts; (5) monitor logs and be ready to revoke the skill if it reads sensitive data or behaves too aggressively. If you can, ask the publisher for a narrower trigger set or explicit constraints on allowed tool usage.Like a lobster shell, security has layers — review code before you run it.
latestvk97ex3q9h2kd2qr8cxa3ky6q6582r8r8
License
MIT-0
Free to use, modify, and redistribute. No attribution required.