Turn raw data into decisions with statistical rigor, proper methodology, and awareness of analytical pitfalls.
Security Analysis
high confidenceThe skill is an instruction-only data-analysis guide whose requested resources and instructions align with its stated purpose — no installs, credentials, or out-of-scope actions are present.
Name and description match the included SKILL.md, pitfalls.md, and techniques.md content. All material is about statistical methodology and analysis practices; there are no unrelated dependencies, credentials, or capabilities requested.
Runtime instructions are procedural guidance for how to perform analysis (questions to ask, checklists, methods to use). They do not instruct the agent to read specific system files, access environment variables, call external endpoints, or exfiltrate data. The scope is limited to analysis methodology.
No install spec and no code files — the skill is instruction-only, so nothing is written to disk or fetched at install time.
The skill requires no environment variables, credentials, or config paths. Nothing asks for secrets or unrelated service access; requested permissions are proportionate to being a guidance-only skill.
always is false and there is no self-modifying install behavior. The skill does not request persistent system presence or access to other skills' config. Autonomous invocation is permitted (platform default) but does not combine with other risky factors here.
Guidance
This skill is a text-only guide for statistical analysis and appears internally consistent. Before using it, remember: it does not itself access your files or systems — it will analyze only the data you provide to the agent. Avoid pasting sensitive PII or secrets into prompts. If your agent has connectors (databases, cloud storage), verify those connectors' permissions separately before allowing the agent to fetch data. Review the pitfalls.md and techniques.md recommendations (sample size, multiple comparisons, confounders) and treat the skill's outputs as methodological guidance you should validate — e.g., rerun tests or check assumptions with your own tooling.
Latest Release
v1.0.0
Initial release
More by @ivangdavila
Published by @ivangdavila on ClawHub
