Fitbit Insights

ReviewAudited by ClawScan on May 10, 2026.

Overview

This Fitbit skill has a clear purpose, but it asks for long-lived Fitbit credentials while the package lacks the scripts it claims to run and its privacy wording is overstated.

Review before installing. Verify the missing Python scripts from a trusted source, make sure the Fitbit app uses read-only scopes, protect `/root/clawd/fitbit-config.json`, and do not use the skill if you are not comfortable giving the agent access to your Fitbit health data.

Findings (4)

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.

What this means

The agent could access sensitive Fitbit activity, heart rate, sleep, and profile data for as long as the refresh credentials remain valid.

Why it was flagged

The skill stores Fitbit OAuth credentials locally, including a client secret and refresh token, while the registry metadata declares no credential or required config path.

Skill content
"client_secret": "YOUR_CLIENT_SECRET", "access_token": "YOUR_ACCESS_TOKEN", "refresh_token": "YOUR_REFRESH_TOKEN" ... Location: `/root/clawd/fitbit-config.json`
Recommendation

Install only if you trust the source, use read-only Fitbit scopes, restrict permissions on the config file, and revoke the Fitbit app tokens if you stop using the skill.

What this means

The described credential-handling and API behavior cannot be reviewed from the package, and the skill may fail or depend on unreviewed scripts outside the submitted artifacts.

Why it was flagged

The documentation claims executable helper scripts are included, but the provided manifest contains only markdown files and the scan says there are no code files to analyze.

Skill content
Inside skill: ... fitbit_api.py - API client ... refresh_token.py - Auto-refresh system
Recommendation

Require the actual `fitbit_api.py` and `refresh_token.py` sources to be included and reviewed before trusting the skill with OAuth tokens.

What this means

A user may believe their health data never leaves Fitbit, when the skill is designed to retrieve it for AI-generated answers.

Why it was flagged

The same artifacts say the skill fetches Fitbit data and has the AI analyze it, so this privacy claim understates that health data enters the assistant/messaging context.

Skill content
Your data - Stays in your Fitbit account, never shared
Recommendation

Clarify exactly where Fitbit data is processed, displayed, logged, or retained, especially when used through messaging integrations.

What this means

Your fitness, sleep, and heart-rate metrics may be included in assistant responses or conversation history.

Why it was flagged

The Fitbit provider flow is purpose-aligned and disclosed, but it brings sensitive health data into the agent's working context.

Skill content
Data Sources: Activity metrics ... Heart rate data ... Sleep tracking ... Workout logs ... Historical trends
Recommendation

Use the skill only for questions you are comfortable exposing to the assistant, and confirm any logging or retention behavior in the host environment.