Ravi login
WarnAudited by ClawScan on May 10, 2026.
Overview
The skill matches its stated Ravi login purpose, but it gives the agent broad access to stored credentials, OTPs, and inbox verification data with weak per-service scoping.
Install or use this only if you trust the Ravi CLI/service and want an agent to sign up for or log into accounts for you. Confirm the exact website/domain before account creation or login, require scoped OTP/email matching, and understand that Ravi auth keys and passwords may be stored persistently.
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.
If invoked in the wrong context, the agent could access or use saved credentials and Ravi identity material to log into accounts.
The skill instructs the agent to rely on persistent local Ravi auth keys and retrieve full stored website passwords for login. That is expected for this purpose, but it is high-impact credential authority and is not tightly bounded by explicit per-domain approval or metadata-declared credential/config requirements.
Store all keys in `~/.ravi/config.json` ... `CREDS=$(ravi passwords get "$UUID")` ... `PASSWORD=$(echo "$CREDS" | jq -r '.password')`
Use only with explicit user approval for the target service/domain, avoid exposing passwords in conversation, and document the Ravi config and credential scope clearly.
The agent could use the wrong OTP or verification link, or surface unrelated private inbox content during a login workflow.
The examples read broad inbox data and take the first numeric SMS code or first email thread, without verifying the sender, service, domain, or request context.
CODE=$(ravi inbox sms | jq -r '.[].preview' | grep -oE '[0-9]{4,8}' | head -1)
THREAD_ID=$(ravi inbox email | jq -r '.[0].thread_id')Filter OTPs and verification emails by the intended service, sender, domain, timestamp, and user confirmation before using them.
The skill may fail without the CLI, or users may run an unreviewed external tool to perform sensitive authentication work.
The skill depends on an external `ravi` CLI, while the supplied artifact set has no install spec, no code files, and no required-binary declaration. This is purpose-aligned but leaves the CLI provenance outside this review.
Before doing anything, check whether you're already authenticated: ```bash ravi auth status ```
Declare the required Ravi CLI, its installation source, version expectations, and any local config paths in the skill metadata.
Third-party services may treat the generated identity as a real person or account holder.
The skill intentionally uses a Ravi-generated identity for third-party forms. This is disclosed and aligned with the skill, but users should understand the trust and policy implications of using a realistic synthetic identity.
The server auto-generates a realistic human name (e.g. "Sarah Johnson") and matching email. ... **Never** use the account owner's name for form fields.
Use this only where a separate Ravi identity is permitted, and make sure the user understands which identity is being presented.