Greptile

v1.0.0

Greptile integration. Manage data, records, and automate workflows. Use when the user wants to interact with Greptile data.

0· 66·0 current·0 all-time
byVlad Ursul@gora050
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
The SKILL.md describes connecting to Greptile via Membrane and shows how to discover and run actions; required steps (install Membrane CLI, login, create a connection, run actions or proxy requests) match the stated purpose.
Instruction Scope
Instructions are limited to installing/using the Membrane CLI, performing login, creating connections, listing actions, running actions, and proxying API calls through Membrane. The skill does not instruct reading unrelated files or exporting arbitrary local data.
Install Mechanism
The guide tells users to install a global npm package (@membranehq/cli). That is expected for a CLI-based integration, but npm packages carry supply-chain risk—users should verify the package and its publisher (npm/GitHub) before installing globally.
Credentials
No environment variables or local secrets are requested. Authentication is delegated to Membrane via an interactive login flow, which is proportional to the tool's needs and avoids asking for direct API keys in the skill.
Persistence & Privilege
The skill is instruction-only, does not request always:true, and does not ask to modify other skills or system-wide settings. Normal agent autonomy would allow invoking these CLI commands only if the runtime grants shell/command access.
Assessment
This skill appears coherent: it delegates auth to Membrane and shows how to use the Membrane CLI to talk to Greptile. Before installing or running anything, verify the @membranehq/cli package and its publisher on npm/GitHub, avoid running global installs as root if possible, and ensure you trust getmembrane.com and the Membrane account flow. If you run this inside an automated agent environment, confirm whether that environment is allowed to run CLI commands and open browser logins (or use the documented headless flow).

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

latestvk97c133q6vx6s3kby70s4k3pk5844wax

License

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

Comments