Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Telegram Compose
v1.0.4Format and deliver rich Telegram messages with HTML formatting via direct Telegram API. Auto-invoked by the main session for substantive Telegram output — no other skills need to call it. Decision rule: If your Telegram reply is >3 lines or contains structured data (lists, stats, sections, reports), spawn this as a Haiku sub-agent to format and send. Short replies (<3 lines) go directly via OpenClaw message tool. Handles: research summaries, alerts, status updates, reports, briefings, notifications — anything with visual hierarchy.
⭐ 4· 4k·14 current·15 all-time
byTrevin@tmchow
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
medium confidencePurpose & Capability
The skill's name and description match what the SKILL.md instructs: read a specified bot token from the OpenClaw config and send formatted Telegram messages via api.telegram.org. One inconsistency: the registry metadata lists no required binaries, while the SKILL.md metadata and examples require jq and curl. Needing the bot token (channels.telegram.accounts.<account>.botToken) is appropriate for the stated purpose.
Instruction Scope
Instructions are narrowly scoped to formatting content and using a provided account's botToken from the local OpenClaw config to call Telegram's sendMessage API. The skill explicitly forbids auto-selecting or iterating accounts (caller must provide account name). It uses shell examples (jq/curl) and requires Read/exec tools; these are expected for this task but grant the sub-agent the ability to run commands and read local files — so correct behavior depends on the agent obeying the 'do not auto-select' rule. The SKILL.md also directs returning message_id but not the message contents, which is coherent.
Install Mechanism
Instruction-only skill with no install spec and no code files — lowest-risk delivery model. Nothing is downloaded or written by an installer.
Credentials
The only sensitive access required is the Telegram bot token stored in the OpenClaw config at channels.telegram.accounts.<account>.botToken, which is proportional to sending messages. No unrelated credentials or environment variables are requested. The only minor issue is the documentation mismatch about required binaries (jq/curl) vs registry metadata.
Persistence & Privilege
The skill is not marked always:true and does not request persistent system-level privileges or modify other skills. It is designed to be spawned as a sub-agent when needed and to read only the specified OpenClaw config path for the provided account.
Assessment
This skill appears to do what it says: format messages and call Telegram's API using a bot token stored in your OpenClaw config. Before installing, confirm: (1) jq and curl are actually available on hosts that will run this skill (SKILL.md uses them, even though the registry metadata omitted them), (2) the OpenClaw config path (~/.openclaw/openclaw.json or ~/.openclaw/clawdbot.json) contains only the bot tokens you expect and has proper filesystem permissions, (3) callers must supply the account name — verify your agent enforces that rule (the skill's safety depends on not auto-iterating accounts), and (4) evaluate whether allowing the sub-agent exec/read tools matches your risk posture (these tools enable the shell examples but could run arbitrary commands if the agent is compromised). If uncertain, test with a throwaway Telegram bot/account and rotate tokens after testing.Like a lobster shell, security has layers — review code before you run it.
latestvk974y7vsbsnt3nsamqsmsaqdks811crv
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
OSmacOS · Linux
Configchannels.telegram.accounts.<account>.botToken