Phy Telegram Bot Payments
Security checks across static analysis, malware telemetry, and agentic risk
Overview
The skill's SKILL.md implements a reasonable paywall flow, but the manifest omits required environment variables and the instructions ask the agent to read/write workspace files and trust client-supplied IDs — these mismatches and trust assumptions are concerning and should be fixed or reviewed before use.
What to consider before installing: - Manifest mismatch: The registry metadata declares no required env vars, but the SKILL.md expects TELEGRAM_BOT_TOKEN, STRIPE_SECRET_KEY, and STRIPE_WEBHOOK_SECRET (and optionally WORKSPACES_DIR). Ask the author to update the manifest to list these required credentials so you can review and approve them explicitly. - Protect secrets: If you proceed, store TELEGRAM_BOT_TOKEN and Stripe secrets in a secure secrets store, not plaintext. Ensure STRIPE_WEBHOOK_SECRET is set and that the webhook verification (Stripe signature) is enforced. - Restrict filesystem access: The code writes to /workspaces/{USER_ID}/usage.json. Ensure WORKSPACES_DIR is configured to a directory dedicated to this bot (not a shared system path) and that file permissions prevent cross-user access. - Avoid trusting client-supplied IDs: The SKILL.md suggests appending client_reference_id to Stripe links. That can be spoofed by a user crafting links. Prefer creating per-user Stripe Checkout sessions server-side or verify a session/charge with Stripe API before crediting. - Review the full implementation: SKILL.md is truncated here — request the complete runtime code/instructions and review that webhook endpoints verify signatures, that credits cannot be arbitrarily injected, and that outgoing Telegram calls only use the intended bot token and chat IDs. - Run in isolation and test: Deploy behind HTTPS (required by Stripe/Telegram webhooks), run in a test environment with test Stripe keys first, and monitor for unexpected writes or messages. Rotate tokens if anything looks suspicious. - If you lack internal expertise, ask a developer or security engineer to review the webhook verification and workspace path configuration before enabling this skill in production.
Static analysis
No static analysis findings were reported for this release.
VirusTotal
No VirusTotal findings
Risk analysis
No visible risk-analysis findings were reported for this release.