Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add sh0tn1k/telegram-ads/plugin install telegram-adsWrote 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.
[](https://agentmods.dev/skills/sh0tn1k/telegram-ads/install-hermes-telegram-ads-watcher)<a href="https://agentmods.dev/skills/sh0tn1k/telegram-ads/install-hermes-telegram-ads-watcher"><img src="https://agentmods.dev/badge/skills/sh0tn1k/telegram-ads/install-hermes-telegram-ads-watcher/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.
<a href="https://agentmods.dev/skills/sh0tn1k/telegram-ads/install-hermes-telegram-ads-watcher"><img src="https://agentmods.dev/badge/skills/sh0tn1k/telegram-ads/install-hermes-telegram-ads-watcher.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00071 | $0.05748 |
| Opus 5 | $0.00036 | $0.02874 |
| Sonnet 5 | $0.00014 | $0.01150 |
| Haiku 4.5 | $0.00007 | $0.00575 |
Grade A, and why
install-hermes-telegram-ads-watcher scanned grade A with 1 finding 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 10d ago.
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.
Runs shell commandslowCapability
Expected in a hook, worth knowing in a rule or an instructions file.
- **Operator activation via systemd env vars — `_global_lifecycle` is never lazy-initialised.** Setting `TELEGRAM_ADS_OPERATOR_ENABLED=1` + `TELEGRAM_ADS_OPERATOR_POLLING_ENABLED=1` in the systemd override is necessary b How it starts
The opening of the file, as written. The whole thing — 182 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Install hermes_telegram_ads.watcher (pinned) into Hermes
When to use
- Нужно подтянуть изменения watcher из upstream
example/telegram-ads-upstreamна pinned commit. - Нужно проверить
import hermes_telegram_ads.watcher+list_tool_coverage()(~62 capabilities). - Нужно подготовить Hermes wiring (store / service / scheduler / read-only adapter / consumer stub) без запуска против реального Telegram Ads кабинета.
- Нужен one-shot real adapter smoke (
detect_login_state/browser_healthy/list_accounts) — отдельный approval gate, не выполняется автоматически. - Нужен долгий
WatcherScheduler.run_forever()background process — отдельный approval gate.
When NOT to use
- Любая задача, которая в итоге вызовет
WatcherScheduler.run_forever()против реального кабинета — это требует отдельного явного одобрения оператора. - Mutating Telegram Ads actions (create/edit/stop/CPM/budget/apply_approved_action) — запрещены в этом skill.
- Запуск
real_adapter_smoke.pyбез явного одобрения оператора на конкретный список вызовов (smoke сам по себе read-only, но инициирует real Chromium launch + GET к ads.telegram.org). - Watches of any kind — adding a watch is a separate approval gate that transitions the scheduler from idle to active (real adapter calls per tick).
Strict approval-scope discipline (HARD rule)
When the operator approves an approval-gated operation with an explicit list of allowed commands/tools, execute ONLY those — nothing else.
This is non-negotiable even when extra read-only tool calls would be helpful for the report:
- ✅ Allowed: one
telegram_ads_login_check()→ report based on its return value alone. - ❌ NOT allowed even if read-only: "while I'm at it, let me also call
telegram_ads_statusandtelegram_ads_get_browser_profile_infofor cross-verification."
Why: the operator treats every additional tool call as a separate side-effect surface. Even a read-only telegram_ads_* call initiates Chromium IPC, navigates to ads.telegram.org, and consumes a request slot — each call is a separate action with its own risk profile. "Helpful cross-verification" is scope drift and will be flagged.
What ships with it
16 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/approval-scope-discipline.md 3.6 KB
- references/browser-lock-parent-attribution.md 6.4 KB
- references/config-profile-path-divergence.md 9.3 KB
- references/login-state-scheduler-gate.md 7.2 KB
- references/metrics-adapter-regression.md 1.5 KB
- references/operational-readonly-policy.md 3.2 KB
- references/operator-activation.md 5.5 KB
- references/real-adapter-smoke.md 5.9 KB
- references/restart-notification-and-guard.md 1.7 KB
- references/watcher-api-and-gotchas.md 6.0 KB
- scripts/ads_watcher_smoke.py 5.1 KB runs code
- scripts/create-restart-marker.sh 564 B runs code
- templates/real_adapter_smoke.py 5.9 KB runs code
- templates/start_ads_watcher_real_login_only.py 3.4 KB runs code
- templates/start_ads_watcher.py 2.1 KB runs code
- templates/test_config_loader_consistency.py 5.5 KB runs code
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.
- 10d ago First seen · 182 lines · 71 tokens per session scan A 355c7f5b37f3
install-hermes-telegram-ads-watcher is a skill published in the GitHub repository sh0tn1k/telegram-ads (0 stars, last pushed 10d ago), licensed MIT. It adds 71 tokens to every session and 5,748 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…