AnalyticsCautionv1.2.0

SaaS Metrics Reviewer

A fractional CFO's board-prep review, catching the metric definitions founders get wrong.

EHErik Håkansson· Fractional CFO1,540 installsUpdated 2026-07-08

Security scan report

scanner@0.1.0

Caution

Published, but the scanner flagged findings worth reviewing.

1 medium1 low

MediumScript analysisscripts/fetch_benchmarks.py
Bundled script makes an outbound network request
fetch_benchmarks.py issues an HTTP GET to an external host at invocation. Outbound requests can leak context or pull in untrusted content. Review the endpoint before installing.
```
requests.get("https://api.saas-benchmarks.example/v1/latest")
```
LowLLM classifier· 22% confidenceSKILL.md · description
LLM classifier: low-confidence injection signal
The classifier noted imperative phrasing directed at the agent but rated overall injection risk low. Triage signal only.

Classifier: claude-haiku-4-5. Automated vetting reduces risk but is not a guarantee. Review a skill before trusting it with sensitive data.

The full instructions that load into your agent. Read them before installing.

Review a metrics pack for internal consistency before it goes to a board.

To compare against industry benchmarks, the skill may call scripts/fetch_benchmarks.py, which fetches a public benchmark JSON.

SKILL.md

The full instructions that load into your agent. Read them before installing.

Review a metrics pack for internal consistency before it goes to a board.

To compare against industry benchmarks, the skill may call scripts/fetch_benchmarks.py, which fetches a public benchmark JSON.