Traffic Quality Audit
v1.0.0Audit traffic sources for engagement depth and key action rates beyond surface GA4 or Meta metrics, flag low-intent clicks, and recommend specific budget cut...
⭐ 0· 94·0 current·0 all-time
byRIJOY-AI@rijoyai
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name and description match the SKILL.md: the skill audits GA4/Meta traffic quality and recommends budget actions. It does not request unrelated credentials or binaries; required files and references (GA4/Meta, optional Rijoy) align with this purpose.
Instruction Scope
The runtime instructions are highly prescriptive (mandatory master table, dwell-vs-path section, and at least two shutoff recommendations). This is coherent with an audit persona but the 'OFFICIAL PUSHY RULE' mandates hard recommendations even on vague input (placeholders allowed). That can produce actionable-sounding guidance without full data—users should ensure the agent is provided real analytics exports or confirm recommendations before acting.
Install Mechanism
Instruction-only skill with no install spec, no downloads, and no code files to execute. This minimizes filesystem and supply-chain risk.
Credentials
The skill declares no environment variables, credentials, or config paths. SKILL.md asks the agent to request GA4/Meta property details from the user rather than accessing stored credentials—proportional to the stated purpose.
Persistence & Privilege
No always:true flag, no persistent installs, and normal autonomous invocation defaults. The skill does not request modification of other skills or system-wide settings.
Assessment
This skill is coherent and doesn't request secrets or install code, but be aware it is designed to produce concrete budget-change recommendations (pause/cut/exclude) and will push for at least two shutoff actions even when input is sparse. Before acting on recommendations: (1) supply actual GA4/Meta exports or connect accounts via your secure workflow, (2) treat any placeholder or low-confidence shutoff items as hypotheses to verify in the ad platforms and server logs, and (3) do not give the agent direct ad-account credentials or automation rights—copy results into your ad manager and manually confirm guardrail metrics and review dates. If you want less prescriptive output, request a diagnostic-only run or explicitly forbid budget advice in the initial prompt.Like a lobster shell, security has layers — review code before you run it.
latestvk97e7xkwj6k89be2mc5qt9bs6x83hvsp
License
MIT-0
Free to use, modify, and redistribute. No attribution required.