Job ID: 6595ba24afca
Run Time: 2026-06-21 16:11:26
Schedule: 0 16 0
[IMPORTANT: The user has invoked the "news-briefings" skill, indicating they want you to follow its instructions. The full skill content is loaded below.]
name: news-briefings
description: "Research current events and deliver formatted news briefings to the user's messaging platform (Telegram/WhatsApp). Covers one-shot news queries and recurring cron-based news briefings — aggregating from RSS feeds via curl, structuring summaries by region/topic, and delivering to the user's preferred channel."
version: 1.3.0
metadata:
hermes:
tags: [news, current-events, rss, briefing, delivery, telegram, whatsapp, cron]
prerequisites: {}
Research and deliver current-events briefings to the user's messaging platform. Supports two modes:
When the user asks for news/current events, gather from multiple RSS feeds using curl (browser dependency may be unavailable). Good free RSS sources:
BBC World: https://feeds.bbci.co.uk/news/world/rss.xml
NYT World: https://rss.nytimes.com/services/xml/rss/nyt/World.xml
Le Monde: https://www.lemonde.fr/international/rss_full.xml
France Info: https://www.francetvinfo.fr/monde.rss (⚠️ /monde/rss → 404, utiliser .rss)
France 24: https://www.france24.com/fr/rss
Hacker News: https://news.ycombinator.com/
Apple Newsroom: https://www.apple.com/newsroom/
The Guardian: https://www.theguardian.com/tone/news
⚠️ Avoid sources with aggressive anti-bot protection — Some sites (Reuters/DataDome, Google News, DuckDuckGo, Bing, Searx) are systematically blocked from certain server IPs. When curl hits these, the connection may get a TCP reset that propagates as a "Broken pipe" (Errno 32) crash, killing the whole agent run — far worse than a simple HTTP error. For one-shot queries, if these fail, just skip them. For cron jobs, remove them from the prompt entirely — a runtime error crashes the agent, and "skip and continue" won't help if curl itself segfaults on a blocked domain.
Use delegate_task for heavier research (multiple feeds → extraction → structured summary) so the main thread stays responsive. Pass toolsets=["terminal"] for simple RSS parsing.
For parallel research on multiple topics, use the tasks= array form:
delegate_task(tasks=[
{"goal": "Research {topic A}", "toolsets": ["terminal"]},
{"goal": "Research {topic B}", "toolsets": ["terminal"]},
])
This runs up to 3 subagents concurrently.
Pitfall: delegate_task is for ONE-SHOT queries only. Do NOT use it in cron job prompts — cron agents cannot use delegate_task.
Format with clear sections by region/topic. Use emoji markers per region:
Each section: 2-3 bullet points max. Bold the headline item (not the emoji). Include source attribution at end.
Check the user's preferred delivery platform:
The send_message(action='list') may not show platforms even when they're configured. If the platform is defined in .env (TELEGRAM_BOT_TOKEN, etc.), try send_message(target='telegram') directly — it routes to the home channel.
Set up automated daily/weekly briefings that deliver directly to the user's messaging platform on a schedule.
Use cronjob(action='create') with:
skills=["news-briefings"] to load this skill automatically"0 16 " for daily at 16h, "0 16 * 0" for Sundays"origin" (default) — auto-routes to the conversation's origin platform (Telegram/WhatsApp if the conversation came from there, CLI otherwise)"telegram:CHAT_ID" — explicit Telegram delivery (e.g., "telegram:8818654814")"local" — no delivery, saved locally"whatsapp:PHONE" — explicit WhatsApp deliveryExample with explicit Telegram delivery:
cronjob(
action='create',
name='Veille quotidienne d\'actualités',
prompt="Génère un briefing quotidien en français...",
schedule='0 16 * * *',
skills=['news-briefings'],
deliver='telegram:8818654814',
)
delegate_task — cron sessions cannot spawn subagentsexecute_code — cron sessions cannot use execute_code (blocked for security). Use write_file or terminal for file operations instead.clarify — cron agents cannot ask the user questionssend_message needed — the job's final response auto-delivers to the target platformIf cronjob(action='list') shows the job as state: scheduled but last_run_at stays null even after next_run_at has passed:
hermes gateway start or hermes gateway installhermes gateway status — if not running, the scheduler is not activeEven feeds classified as "reliable" (BBC, Le Monde, France 24) can intermittently trigger a Broken pipe (Errno 32) crash in cron context. This happens when a CDN or reverse proxy intermittently sends a TCP reset — the curl subprocess crashes before any "skip on error" logic runs.
no_agent=False, script=path) that curl-fetches feeds sequentially in a single bash script with || true fallbacks, then let the LLM summarize the collected data. This isolates the crash risk to the script layer.delegate_task(tasks=[...], toolsets=["terminal"]). Fetch the feeds in parallel subagents (max 3), each handles its own curl failure gracefully. Then compose and deliver the briefing manually. This bypasses the Broken pipe crash because each subagent's curl crash is isolated.last_status: "error" with delivery error about "cannot schedule new futures"When last_status: "error" AND the delivery error contains "cannot schedule new futures after interpreter shutdown" or "Broken pipe" appears in the output log:
curl to a blocked domain (DataDome/Cloudflare) causes a TCP reset → "Broken pipe" (Errno 32)execute_code in cron)~/.hermes/cron/output/<job_id>/ for the crash reason. Fix the prompt:execute_code calls with write_file or terminalcronjob(action='run', job_id='...') to test the fixWhen a cron job shows last_status: "error" but last_delivery_error: null:
cronjob(action='run', job_id='...') to re-kick the job manuallyEach cron briefing can also save a copy locally for historical reference, alongside Telegram delivery.
Pattern: Add an instruction at the end of the cron prompt to write the briefing content to a file:
❗ À la fin, archive le briefing dans un fichier : écris le contenu exact du briefing dans /home/sebastien/.hermes/memory/modules/veille_actualite/daily/YYYY-MM-DD.md (remplace YYYY-MM-DD par la date du jour). Crée le fichier avec la commande d'écriture appropriée.
Use a dedicated module structure under ~/.hermes/memory/modules/:
veille_actualite/daily/ — for daily briefingsveille_actualite/weekly/ — for weekly geopolitics briefingsPre-create the directories before setting up the cron — the cron agent has file-write tools but directory creation is more reliable when done ahead of time:
mkdir -p ~/.hermes/memory/modules/veille_actualite/{daily,weekly}
Updating existing cron jobs to add archiving: use cronjob(action='update', job_id='JOB_ID', prompt='...') with the original prompt + the archive instruction appended. Always read the existing prompt from ~/.hermes/cron/jobs.json first so you preserve the full content.
| Frequency | Cron expression |
|---|---|
| Daily at 16h | 0 16 * |
| Daily at 9h | 0 9 * |
| Weekly Sunday 16h | 0 16 0 |
| Every 2 hours | every 2h |
curl for RSS feeds over browser toolssend_message(action='list') can return empty even with configured platforms — try direct targetingdelegate_task with toolsets=["terminal"] rather than doing curl calls in the main thread — keeps context cleanhttps://www.france24.com/fr/rss) is a reliable French-language source for international news — include it for francophone briefings import urllib.request, json
url = f"https://api.telegram.org/bot{TOKEN}/getMe"
resp = urllib.request.urlopen(url)
data = json.loads(resp.read()) # {"ok":true, "result":{...}}
.env via terminal heredoc may corrupt them (redaction system replaces characters). Always verify with xxd after writing. Use printf '\xHH' or direct Python binary writes to bypass redactionThe skill ships with a reusable data-collection script in scripts/collecte-veille.sh.
Purpose: Eliminates the Broken pipe (Errno 32) crash that kills cron briefings when a CDN sends a TCP reset during curl. Each feed is fetched with || true isolation — if one feed crashes, the others continue unaffected.
Usage in cron: Set script='collecte-veille.sh' on the cron job. The script runs first, its stdout is injected into the cron agent's prompt, and the agent summarizes the pre-collected data instead of running its own curls. The prompt should explicitly say: "Utilise les données collectées — ne fais PAS de curl toi-même."
Feeds collected: BBC World, Le Monde International, France Info (via .rss, not /rss), France 24, The Guardian (HTML topic page). Parses 10 items each.
Customization: Modify the script to add/remove feeds, or adjust the item count per feed.
blogwatcher — for ongoing/monitoring RSS feeds (complementary: use for single-feed watching, use this skill for multi-source LLM-synthesised briefings)editorial-review-cycle — for producing structured daily/weekly reviews that update the editorial memory system (topics, hypotheses, briefs archive, index). Use this skill when the user asks for a "revue quotidienne" or "briefing mémoire"; use news-briefings when they ask for "what's happening in the news" to deliver to a messaging platform.references/rss-feeds.md for the full list of tested RSS endpoints and parsing recipes.references/vps-feed-reliability.md for per-server feed status (which feeds work, which are blocked, intermittent failure patterns).The user has provided the following instruction alongside the skill invocation: [IMPORTANT: You are running as a scheduled cron job. DELIVERY: Your final response will be automatically delivered to the user — do NOT use send_message or try to deliver the output yourself. Just produce your report/output as your final response and the system handles the rest. SILENT: If there is genuinely nothing new to report, respond with exactly "[SILENT]" (nothing else) to suppress delivery. Never combine [SILENT] with content — either report your findings normally, or say [SILENT] and nothing more.]
Génère une veille géopolitique hebdomadaire en français pour Sébastien, couvrant les événements les plus marquants de la semaine.
Utilise curl pour rassembler les infos depuis plusieurs flux RSS :
Structure la veille par grandes régions :
🌍 Tendances globales — conflits, alliances, diplomatie
🔴 Moyen-Orient — Israël/Gaza/Liban, Iran, Syrie
🟠 Ukraine / Russie — front, diplomatie, impacts
🟡 Amériques — États-Unis, Amérique latine
🔵 Asie-Pacifique — Chine, Taïwan, Corée, Indo-Pacifique
🟢 Afrique — Sahel, Grands Lacs, Corne
🟣 Europe / UE — institutions, élections, énergie
🇫🇷 France — politique étrangère, défense, OTAN
Pour chaque section : 2-3 puces maximum avec une synthèse des faits marquants de la semaine. Mets en gras les faits les plus importants. Ajoute une section "📈 À surveiller la semaine prochaine" en fin.
Livré sur Telegram à Sébastien. Français uniquement.
❗ À la fin, archive le briefing dans un fichier : écris le contenu exact du briefing dans /home/sebastien/.hermes/memory/modules/revue_editoriale/briefs/weekly/YYYY-MM-DD.md (remplace YYYY-MM-DD par la date du jour). Crée le fichier avec la commande d'écriture appropriée.
📅 Dimanche 21 juin 2026 — Synthèse des flux BBC World, Le Monde International, France Info et France 24.
🌍 Tendances globales
🔴 Moyen-Orient
🟠 Ukraine / Russie
🟡 Amériques
🔵 Asie-Pacifique
🟢 Afrique
🟣 Europe / UE
🇫🇷 France
📈 À surveiller la semaine prochaine
Sources : BBC World, Le Monde International, France Info Monde, France 24. Synthèse automatique du 21 juin 2026 — vérifier les sources pour tout usage décisionnel. Reuters exclu (bloqué par DataDome sur ce serveur).
📁 Archive locale : /home/sebastien/.hermes/memory/modules/revue_editoriale/briefs/weekly/2026-06-21.md