Explain the collected threads

The aggregation node collects the incoming content fields into an array. The first LLM chain serializes that array into its prompt and asks for themes, shared topics, and why posts use #damus. Its generated text is sent to three branches: an email formatter, a theme-list extractor, and Telegram.

The model output is an interpretation of supplied posts. Check it against the original content before treating a claimed theme or motivation as evidence. The source does not include a citation checker or a human approval gate.

Extract a theme list

The second LLM chain receives the first chain's generated text and asks for a list of themes without a preamble. It therefore summarizes an earlier model response, rather than reading the original posts again at this step. Errors introduced upstream can carry into this list.

The list goes to input 0 of the Merge node. The aggregated source content goes to input 1. The Merge is configured to combine items by position; confirm the item counts and ordering before changing either branch.

Build the detailed report

After the positional merge, the final LLM chain receives both the extracted themes and the aggregated post content. Its prompt requests an overall theme, recurring topics, examples, observations, and possible improvements. The resulting text branches to Gmail and Telegram.

The prompt also asks about user motivation. That is an inference, not an observed fact. Review quotations and examples against the supplied posts, and keep uncertain interpretations labeled. No automated factual-verification node is present in this export.

Collect Nostr posts and configure the entry point

The scheduled entry point connects to the community Nostr reader, configured with strategy hashtag, hashtag #damus, and a from value of 180. Confirm the meaning and units of that last setting in the installed community node; this export does not explain them. Install the compatible n8n-nodes-nostrobots package before testing.

The export also contains a manual trigger, but its connection-map key uses literal escaped quote characters that do not match the actual node name. Reconnect the manual trigger to the Nostr reader in your copy before relying on a manual run. Inspect the schedule interval as well; the exported interval contains no explicit values.

Community node source: https://github.com/ocknamo/n8n-nodes-nostrobots

Send the theme summary to Telegram

This branch takes the first LLM chain's text, keeps only its first 4,000 characters, and sends it to the configured Telegram chat. Select your Telegram credentials and supply your own chat identifier; the export reads it from TELEGRAM_CHAT_ID in the environment.

The node requests HTML parsing while the LLM output is not constrained to valid Telegram HTML. Test formatting with a harmless destination. The character slice can omit part of a summary; it does not split a long message into multiple deliveries.

Deliver the final report

The final report is delivered along two independent branches. One converts the model's Markdown into HTML and passes the resulting data field to Gmail. The other sends the first 4,000 characters of the report directly to Telegram with HTML parsing enabled.

Replace the Gmail recipient placeholder, select your own account credentials, and configure the Telegram destination. Test both branches separately: successful email formatting does not verify Telegram markup or prove that the complete report fits in one message. These nodes send immediately; there is no approval step in the exported graph.

Test with a small controlled sample

Import the workflow inactive, repair the manual-trigger connection, and select a currently available model in each of the three Gemini model nodes. The export names a preview model; this review does not establish its present availability. Configure Gmail, Telegram, and Gemini credentials outside the shared JSON.

Use a small synthetic or non-sensitive post sample and test destinations. Inspect the aggregated content, first summary, extracted list, merged item, and final report before enabling either delivery branch or the schedule. Include empty input and malformed-content cases. The current workflow has not been executed as part of this annotation review.

#️⃣Nostr #damus AI Powered Reporting + Gmail + Telegram

Purpose and structure

This folder contains the #️⃣Nostr #damus AI Powered Reporting + Gmail + Telegram workflow. Its executable nodes use aggregate, chainLlm, gmail, lmChatGoogleGemini, manualTrigger, markdown, merge, nostrobotsread, scheduleTrigger, telegram. The table and connection map below are derived from the JSON, not an inferred guarantee of behavior.

Setup

Import inactive. Configure your own credentials and replace REPLACE_WITH_* values. Review external destinations and Code nodes before running.

Entry points

When clicking ‘Test workflow’, Schedule Trigger

Verification

Sanitized source candidate; not runtime-tested. This overview is generated from node metadata. Read the folder README for the connection map, credential types, and license.

Attribution

https://github.com/Zie619/n8n-workflows/blob/94007c1445d9258a7da116646b79473e7c7c3282/workflows/Telegram/0001_Telegram_Schedule_Automation_Scheduled.json MIT; preserve source notices.