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.
npx skills add ellmos-ai/bach --skill connectorsgit clone --depth 1 https://github.com/ellmos-ai/bachWrote 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/ellmos-ai/bach/connectors)<a href="https://agentmods.dev/skills/ellmos-ai/bach/connectors"><img src="https://agentmods.dev/badge/skills/ellmos-ai/bach/connectors.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 6 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Privilege Escalation · line 49 Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
- high Privilege Escalation · line 59 Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
- high Privilege Escalation · line 73 Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
- high Privilege Escalation · line 140 Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
- high Privilege Escalation · line 151 Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
- high Privilege Escalation · line 152 Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
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 | $0.00041 | $0.01425 |
| Opus 5 | $0.00020 | $0.00713 |
| Sonnet 5 | $0.00008 | $0.00285 |
| Haiku 4.5 | $0.00004 | $0.00143 |
Grade A, and why
connectors 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 7d 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.
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.
How it starts
The opening of the file, as written. The whole thing — 176 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Connectors
Status: BETA
Bidirektionale Kommunikation mit externen Services. 3 Runtime-Adapter verfuegbar, Queue-Processor fuer zuverlaessige Zustellung.
Zwei Subsysteme
1. Volle Connectors (bidirektional)
| Typ | Status | Features |
|---|---|---|
| Telegram | Beta | Polling, Owner-Filter, Send, Poll-Loop |
| Discord | Beta | Webhook-Send, Incremental Polling, Bot-Filter |
| HomeAssistant | Beta | States, Services, History, Events |
| Signal | Geplant | signal-cli Integration |
| Geplant | Baileys Integration |
2. Notification-Channels (nur Senden)
| Channel | Status | Dispatch |
|---|---|---|
| Telegram | Aktiv | Bot API via secrets.json |
| Discord | Aktiv | Webhook |
| Slack | Aktiv | Webhook |
| Webhook | Aktiv | Generischer HTTP POST |
| Aktiv | SMTP_SSL |
Neue Channels: siehe connectors/templates/README.md
Secrets-Management
Keys gehoeren in user/secrets/secrets.json (dist_type=0), NICHT in bach.db!
In der DB steht nur auth_type='secrets_file' als Verweis.
Architektur
CLI / REST-API / Daemon
|
┌──────┴──────────────────────┐
| |
hub/connector.py (bidir.) hub/notify.py (nur Senden)
| |
queue_processor.py _load_secrets()
| |
connectors/ (Adapter) user/secrets/secrets.json
├── base.py (ABC)
├── telegram_connector.py
├── discord_connector.py
└── homeassistant_connector.py
Queue-System (v2.0)
- Retry: Exponentieller Backoff (30s → 480s, 5 Versuche)
- Circuit Breaker: 5 Fehler → 5 Min Sperre, Auto-Reset
- Dead Letter: Nach max_retries, manuell wiederherstellbar
- Kontext-Hints: ContextInjector + context_triggers beim Routing
- Daemon: 2 automatische Jobs (poll_and_route 2min, dispatch 1min)
Verwendung
# Connector registrieren
bach connector add telegram mein_bot
# Einmalig pollen
bach connector poll mein_bot
# Queue verwalten
bach connector queue-status
bach connector retry all
# Automatisch (Daemon)
bach connector setup-daemon
bach daemon start --bg
What ships with it
15 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.
- __init__.py 1.1 KB runs code
- base.py 4.2 KB runs code
- discord_connector.py 11 KB runs code
- homeassistant_connector.py 6.0 KB runs code
- signal_connector.py 8.8 KB runs code
- telegram_connector.py 19 KB runs code
- templates/__init__.py 1.5 KB runs code
- templates/connector_template.py 5.9 KB runs code
- templates/notification_template.yaml 2.9 KB
- templates/QUICKSTART.md 5.4 KB
- templates/README.md 15 KB
- templates/setup_wizard.py 17 KB runs code
- templates/telegram_template.yaml 4.1 KB
- templates/whatsapp_template.yaml 5.2 KB
- whatsapp_connector.py 9.3 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.
- 7d ago First seen · 176 lines · 41 tokens per session scan A 555d135e1d88
connectors is a skill published in the GitHub repository ellmos-ai/bach (7 stars, last pushed 2d ago), licensed MIT. It adds 41 tokens to every session and 1,425 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
pydanticai
Build type-safe AI agents and graph-based workflows with PydanticAI and PydanticGraph. Agent creation, function tools, capabilities, dependency injection, structured output, streaming, multi-agent patterns, testing, evals, and graph state machines. Use whenever you are building agents, tool-using LLM workflows, or…
dev-soft-agent
Automated software development pipeline. Scans projects, prioritizes tasks, analyzes code, and orchestrates development loops. Zero dependencies (Python stdlib only).
knowledge-transfer
Interview a person and turn their operating know-how (tacit knowledge) into the company brain: processes, rules with source, a glossary, decisions and open questions, written into the workspace. Use whenever someone needs to capture how a process actually gets done, before a key person leaves or retires, to onboard a…
senior-strategy-architect
Use proactively whenever the user asks for any strategy, plan, roadmap, growth plan, GTM, marketing, business, product, brand, pricing, sales, content, retention, fundraising, marketplace, partnerships, community, competitive response, launch, monetization, or business decision. Turns vague asks into senior-level…
cowork-os-core
Core operating rules for a cowork-os workspace. Use whenever working inside a Claude Cowork project that uses cowork-os (folders like context/, marketing/, website/, decisions/, signals/, reviews/), or on any strategy, marketing, content, website, or business-development task where decisions and context should persist…
pipeline-followup
Prepare draft follow-up emails for CRM deals that have gone quiet. Use whenever the person wants to chase deals with no reply, re-engage stalled opportunities, or "prepare the follow-ups" for their pipeline — e.g. "controlla la pipeline e preparami i follow-up", "chi non mi ha risposto?", "fammi le bozze per i deal…