Telegram Media Resolver

Resolve Telegram <media:image>, <media:document>, <media:video> and other media placeholders into actual files for vision/analysis. Use when a Telegram message contains a media placeholder (e.g. <media:image>) that you cannot see — typically in quoted/replied-to messages or group chat history. Downloads the media via Telegram Bot API and returns a local file path for the image tool or further processing.

This skill appears to do what it says (download media via the Telegram Bot API), but the documentation instructs the agent to read a local OpenClaw config (~/.openclaw/openclaw.json) to obtain the bot token while the skill metadata declares no required credentials — that mismatch is the main red flag. Before installing or using it: 1) ask the publisher to update the metadata to declare the Telegram bot token as a required credential (or remove the instruction to read the local config); 2) avoid letting the agent automatically read your OpenClaw config—prefer passing the bot token explicitly via a secure input or env var scoped to this skill; 3) review the included script (scripts/fetch_media.py) yourself — it is short and readable and only talks to api.telegram.org — and run it in an isolated environment if you must test; 4) ensure the bot token you provide has minimal permissions (use a dedicated bot) and that the bot is a member of the target chat; and 5) be aware forwarding may briefly appear in chats. If the publisher cannot justify the undeclared config access, treat the skill with caution or refuse to install it.

Type: OpenClaw Skill Name: tg-media-resolve Version: 1.0.0 The skill is designed to resolve Telegram media placeholders by downloading files via the Telegram Bot API. It requires reading the `botToken` from the OpenClaw configuration (`~/.openclaw/openclaw.json`) as instructed in `SKILL.md`, and the `scripts/fetch_media.py` script uses this token to interact with `api.telegram.org` and download files to a specified output directory (defaulting to `/tmp`). While this involves network access, file system writes, and handling a sensitive API token, these actions are directly necessary and aligned with the skill's stated purpose. There is no evidence of intentional harmful behavior such as unauthorized data exfiltration, persistence mechanisms, or broader prompt injection attempts beyond retrieving the required token for its function.