tg-responder

tg-responder is a skill for Claude Code, Codex from glebis/claude-skills. It costs 67 tokens per session (1,005 once invoked), scanned A, original, MIT.

A Telegram assistant that reviews drafted replies and tracks messages that still need a response. Telegram is a messaging service used for private chats, groups, and channels.

In plain words
What is it for?
It is for reviewing response drafts, approving or editing replies, sending them, and finding outbound messages that received no answer.
Why use it?
It reduces the chance of overlooking unanswered messages and keeps proposed replies separate until you approve them.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: reads .claude/ paths.

Good fit It is for reviewing response drafts, approving or editing replies, sending them, and finding outbound messages that received no answer.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/glebis/claude-skills/tg-responder
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add glebis/claude-skills --skill tg-responder
Clone the repo
git clone --depth 1 https://github.com/glebis/claude-skills

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for tg-responder

README.md
[![agentmods](https://agentmods.dev/badge/skills/glebis/claude-skills/tg-responder/github.svg)](https://agentmods.dev/skills/glebis/claude-skills/tg-responder)
Your own site
<a href="https://agentmods.dev/skills/glebis/claude-skills/tg-responder"><img src="https://agentmods.dev/badge/skills/glebis/claude-skills/tg-responder/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for tg-responder

Your own site · 80×15
<a href="https://agentmods.dev/skills/glebis/claude-skills/tg-responder"><img src="https://agentmods.dev/badge/skills/glebis/claude-skills/tg-responder.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 67 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,005 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00067 $0.01005
Opus 5 $0.00034 $0.00502
Sonnet 5 $0.00013 $0.00201
Haiku 4.5 $0.00007 $0.00101

Measured 8d ago against content hash abd6ec7ae39e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

tg-responder scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.

The scan reads SKILL.md. This mod also ships 9 executable files (scripts/backfill.py, scripts/classify.py, scripts/follow_ups.py, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

tg-responder/SKILL.md · 115 lines

How it starts

The opening of the file, as written. The whole thing — 115 lines — stays where its author put it; the contents beside it link to each section on GitHub.

tg-responder — Telegram Communications Assistant

Review pending response drafts and manage the Telegram response queue.

Commands

review — Approve pending drafts

Read the responder queue and present drafts for approval:

python3 ~/.claude/skills/tg-responder/scripts/schema.py  # ensure DB exists

Then query the database:

-- Pending drafts needing approval
SELECT o.id, o.chat_id, o.draft_text, o.draft_reason, o.source,
       i.sender_name, i.text as original_text, i.urgency, i.category,
       datetime(i.received_at, 'unixepoch') as received
FROM outbox o
JOIN inbox i ON o.inbox_id = i.id
WHERE o.status = 'draft'
ORDER BY
  CASE i.urgency WHEN 'urgent' THEN 0 WHEN 'normal' THEN 1 ELSE 2 END,
  o.created_at ASC;

For each draft, present to the user:

  1. Original message — who sent it, when, what they said
  2. Draft response — the proposed reply
  3. Options: approve (send as-is), edit (modify then send), skip

To approve and send a draft:

  1. Update outbox: UPDATE outbox SET status = 'approved', final_text = draft_text, approved_at = strftime('%s','now') WHERE id = ?
  2. Send via telegram skill: python3 ~/.claude/skills/telegram/scripts/telegram_fetch.py send --chat-id CHAT_ID --text "THE_TEXT"
  3. Update outbox with sent status and message_id

To skip: UPDATE outbox SET status = 'skipped', updated_at = strftime('%s','now') WHERE id = ?

status — Queue statistics

-- Inbox stats
SELECT status, count(*) FROM inbox GROUP BY status;

-- Outbox stats
SELECT status, count(*) FROM outbox GROUP BY status;

-- Recent activity
SELECT sender_name, route, status, datetime(created_at, 'unixepoch')
FROM inbox ORDER BY created_at DESC LIMIT 10;

Report: pending count, drafts waiting, sent today, failed items.

follow-ups — Track unanswered outbound messages

Scan for people who haven't replied, send reminders with exponential backoff.

# Scan for new unanswered messages (needs Telethon session — stop daemon first)
python3 ~/.claude/skills/tg-responder/scripts/follow_ups.py scan

# Process due reminders (drafts to Telegram or outbox)
python3 ~/.claude/skills/tg-responder/scripts/follow_ups.py remind

# List active follow-ups
python3 ~/.claude/skills/tg-responder/scripts/follow_ups.py list

# Archive expired follow-ups
python3 ~/.claude/skills/tg-responder/scripts/follow_ups.py archive

# Run all (scan + check replies + remind + archive)
python3 ~/.claude/skills/tg-responder/scripts/follow_ups.py all

Read the full file on GitHub · 115 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 8d ago First seen · 115 lines · 67 tokens per session scan A abd6ec7ae39e

Subscribe to this mod's changes

tg-responder is a skill published in the GitHub repository glebis/claude-skills (375 stars, last pushed 10d ago), licensed MIT. It adds 67 tokens to every session and 1,005 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.

Related

Other skills, from other repositories