GutCheck Digestive Health Tracker

v1.0.0

GutCheck - A digestive health tracking application with personalized insights and data-driven recommendations. Helps users understand food sensitivities and optimize gut wellness.

1· 1.6k·1 current·1 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
medium confidence
!
Purpose & Capability
The skill's name/description promise a full application (server + client) but the bundle only contains docs, a publish script, SKILL.md, and package.json — no server index.js or client/ directory in the provided manifest. The SKILL.md expects cloning https://github.com/openclaw/gutcheck.git and running npm install there; that could provide the app, but the packaged files do not actually contain the app. This mismatch (advertised app vs. provided contents) is incoherent and unexplained.
!
Instruction Scope
Runtime instructions tell the agent to clone a GitHub repo and run npm install (including a client npm install), create a .env with MONGODB_URI and JWT_SECRET, and run the app — yet the registry metadata declared no required env vars and the manifest lacks the referenced client folder and server entry point. SKILL.md also inconsistently lists PORT=5000 but says the app is available at localhost:3000. Instructions therefore reference files/paths/envs not present in the package and request creation of sensitive variables without declaring them.
Install Mechanism
The install plan clones a GitHub repo (well-known host) and runs npm install in the project and client directories. Cloning from GitHub is expected for code installs, but running npm install on an external repo can execute arbitrary postinstall scripts. The install uses standard tools (git, npm) and no obscure URLs, but because the package as provided lacks the app code, the actual behavior depends on the remote repo content — this raises moderate risk until the repo is inspected.
!
Credentials
SKILL.md asks users to create a .env with MONGODB_URI and JWT_SECRET (sensitive credentials) but the registry metadata lists no required env vars or primary credential. Requiring a database URI and JWT secret is plausible for an app, but requesting them without declaring them in metadata and without including the app source in the package is disproportionate and unclear. The skill asks for credentials that could be misused if the upstream code is untrusted.
Persistence & Privilege
The skill is not marked always:true and does not request system-wide persistence or other skills' configs. The publish script creates and removes a temporary directory but does not alter system-wide settings. Autonomous invocation is allowed (normal) but not combined here with any other high-risk privilege.
What to consider before installing
What to consider before installing: - Do not install or run this on a machine with real user data or production credentials until you verify the remote repository contents. - The package you were shown does not include the actual server/client application — it contains marketing docs, SKILL.md, package.json, and a publish script. Confirm that https://github.com/openclaw/gutcheck.git actually contains the expected app source before cloning and running npm install. - SKILL.md asks you to create a .env with MONGODB_URI and JWT_SECRET (sensitive). Only provide these after you have inspected the code and are running in an isolated environment (local dev VM or container) with test data. - npm install on an untrusted repo can run arbitrary code (postinstall scripts). Inspect package.json and any lifecycle scripts in the remote repo before executing. - The skill has small inconsistencies (PORT=5000 vs. documentation saying localhost:3000, directory name mismatches like 'GutCheck' vs. 'gutcheck'), suggesting sloppy packaging — ask the author for the canonical repo URL and an explanation of the missing app code. If you want to proceed safely, request from the publisher: 1) A link to the exact GitHub repo commit/tag that contains the server and client code. 2) A file listing (tree) showing the server entry point (index.js/app.js) and client folder. 3) Confirmation that no postinstall scripts run network calls or exfiltrate data (or provide the package.json from the remote repo for review). Given the mismatches, I rate this skill as suspicious rather than benign; additional verification of the remote code would likely move the confidence higher.

Like a lobster shell, security has layers — review code before you run it.

latestvk9726km7dhxrg3pqdtmmjwswq180emgw

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Runtime requirements

Binsnode, npm

Comments