Approval required

telegram-automation

OpenClawClaudeCodexHermes Agent

Telegram Conditional Reply Workflows

Plan and dry-run Telegram replies based on conversation context, tags, or follow-up state.

Example request

Illustrative scenario, not an executed run.

Dry-run a reply rule for these tagged chats. Show which conversations match and the proposed reply; do not execute it.

Expected result

A dry-run report of matching conversations and proposed replies, so you can check the rule before execution.

Before you start

Connect Telegram and select the personal or team context you intend to use.
Connect your AI client with browser OAuth, or follow the local setup below.
Review the workflow’s tools and target conversations. Chat reads and imported CRM records are separate; sync a chat before requesting changes to missing CRM metadata.

A skill does not schedule itself or grant permissions. Describe whether you want suggestions, a dry run, or an action. Client controls and each tool’s safeguards govern execution.

Flow summary

Parsed from the packaged SKILL.md.

telegram-conditional-replies

Use this skill for rule-driven reply workflows.

Rules

Start with dry-run or recommendation mode unless the user explicitly asks to enable execution.
Prefer CRM tasks or draft replies before automatic sends.
Read recent history before proposing reply conditions.
Use explicit enablement and approval for any rule that sends Telegram messages.
Do not create broad always-on reply rules without a narrow condition and audit path.

Flow

Inspect existing rules with rules_list.
Read target chats with dialogs_list and chat_read when needed.
Add or adjust rule instructions with rules_add if the user requests persistence.
On local tgchats, dry-run the rule with rules_dry_run. Chiho Cloud does not expose that local-only tool, so inspect the proposed conditions and affected scope without running the rule.
Inspect outcomes with rules_log.
Run the rule with rules_run only after the local dry-run or hosted review is acceptable and the user explicitly approves execution.
For actual replies, create a non-sending preview with outbox_preview. After explicit approval, Chiho Cloud must call write_approve_preview and then outbox_send_approved; local tgchats calls outbox_send_approved after the user approves. message_send_draft is a direct-send tool and must only be used when the user explicitly asks to send immediately.

Execution Boundary

Use rules_disable or rules_delete to stop or clean up persistent conditional reply rules. Until rules_update exists, replace a rule by creating a narrower new rule and deleting the old one after review.

First-Time Setup

If Telegram is not connected yet, start with the root Chiho Telegram skill:

https://raw.githubusercontent.com/chihoai/telegram-for-ai-agents/main/SKILL.md

Use that root skill to choose Chiho.ai Cloud or self-hosted tgchats, then return to this workflow skill.

References

Required tool permissions

telegram.read
crm.write
telegram.message.preview
telegram.message.send
telegram.batch.write
automation.rules.write

Use Chiho.ai with a professional team CRM

Choose Chiho.ai when you want:

A hosted CRM table UI for chats, contacts, tags, tasks, and follow-ups.
Browser OAuth connections you can review and revoke.
Shared team context and tool-specific action safeguards.
No local Telegram worker, database, or MCP runtime to operate.

Connect with browser OAuth

Run SKILLs in your personal local CLI

Choose the local setup when you want:

Open-source workflows you can inspect, change, and run yourself.
A local Telegram session with tgchats CLI and tgchats-mcp access.
Control over your database, worker, backups, export, and import path.
A developer-friendly runtime for custom Skills and self-hosted automation.

Setup README

Cloud MCP

Required scopes:

telegram.read
crm.write
telegram.message.preview
telegram.message.send
telegram.batch.write
automation.rules.write

Read/CRM tools:

rules_list
rules_add
rules_run
rules_log
dialogs_list
chat_read

Write tools for message execution:

message_send_draft
outbox_preview
write_approve_preview
outbox_send_approved

Chiho Cloud does not expose the local-only rule dry-run tool. Review the rule conditions and affected scope before calling rules_run. Use outbox_preview for a non-sending message preview, call write_approve_preview only after explicit approval, and then call outbox_send_approved to execute the approved preview. Approval alone does not send. message_send_draft sends directly.

Local tgchats

If auth or any Telegram command prints a QR login code, show the full QR code block and expiry to the user so they can scan it; keep the process running until login completes, 2FA is needed, or the user asks to stop.

Use local tgchats-mcp for rule and read tools. For message execution:

message_send_draft
outbox_preview
outbox_send_approved

Always call rules_dry_run before rules_run for conditional reply workflows. Use outbox_preview for a non-sending preview and call outbox_send_approved only after explicit approval. message_send_draft sends directly.

Templates

Formatted preview from the packaged JSON asset.

Availability check

text

Hi {{first_name}}, thanks for the note. I can take a look and get back to you by {{date}}.

Handoff reply

text

Hi {{first_name}}, looping in {{colleague_name}} who can help with {{topic}}.

Examples

Formatted preview from the packaged JSON asset.

OpenClaw

prompt

Create a rule that drafts replies for chats tagged customer when they ask about pricing.

tools

rules_list

chat_read

rules_add

Codex

prompt

Run the reply rules and show me what would happen before sending anything.

tools

rules_run

rules_log

Claude Desktop

prompt

Draft a conditional reply for this chat, but do not send it yet.

tools

chat_read

outbox_preview

Keep building your workflow

For unattended jobs, review advanced service-token setup separately.