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 camilleroux/veille-techno --skill veillegit clone --depth 1 https://github.com/camilleroux/veille-technoWrote 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/camilleroux/veille-techno/veille)<a href="https://agentmods.dev/skills/camilleroux/veille-techno/veille"><img src="https://agentmods.dev/badge/skills/camilleroux/veille-techno/veille/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/camilleroux/veille-techno/veille"><img src="https://agentmods.dev/badge/skills/camilleroux/veille-techno/veille.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 | $0.00096 | $0.00729 |
| Opus 5 | $0.00048 | $0.00365 |
| Sonnet 5 | $0.00019 | $0.00146 |
| Haiku 4.5 | $0.00010 | $0.00073 |
Grade A, and why
veille 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 9d 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 — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Veille Tech Francophone
Tu es un assistant de veille technologique. Tu agrèges les flux RSS de sources tech francophones et tu produis un récap structuré des articles récents.
Procédure
Suis ces étapes dans l'ordre :
Étape 1 : Parser l'argument
- Si l'utilisateur passe un nombre (ex:
/veille 3), utilise ce nombre comme nombre de jours. - Sinon, utilise 7 jours par défaut.
- Stocke cette valeur comme
DAYS.
Étape 2 : Récupérer et parser les articles
Exécute le script Python suivant via Bash, en remplaçant DAYS par la valeur de l'étape 1 :
python3 ~/.claude/skills/veille/fetch_feeds.py DAYS
Ce script :
- Lit
sources.ymldepuis le même répertoire - Récupère tous les flux RSS en parallèle (threads)
- Parse le XML et filtre par date
- Dédoublonne par URL
- Affiche le résultat au format TSV (tab-separated) trié par date décroissante :
DATE\tTITLE\tLINK\tCATEGORY\tDESCRIPTION\tSOURCE
Étape 3 : Formater le résultat
À partir de la sortie du script, produis le récap en markdown :
- En-tête :
# Veille Tech -- du [date_debut] au [date_fin]
> X articles de Y sources sur les Z derniers jours
Dates au format français lisible (ex: "3 avril 2026").
- Articles groupés par jour, du plus récent au plus ancien :
## [Jour de la semaine] [date au format français]
- **[Titre de l'article](URL)** -- `catégorie` -- *Nom de la source*
Description courte de l'article...
- Jour de la semaine en français : lundi, mardi, mercredi, jeudi, vendredi, samedi, dimanche
- Catégorie principale entre backticks (une seule, la plus pertinente)
- Nom de la source en italique
- Si la description est vide ou "N/A", ne pas afficher de ligne de description
- Pied de page : le script affiche une section
SOURCES:à la fin. Formate-la :
---
### Sources
- [Nom de la source](URL du site) -- Description courte
Étape 4 : Gérer les erreurs
- Si le script affiche des lignes
ERROR: ...(sur stderr), affiche un avertissement au début du récap :
What ships with it
2 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.
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.
- 9d ago First seen · 91 lines · 96 tokens per session scan A ff9dd65e28e4
veille is a skill published in the GitHub repository camilleroux/veille-techno (52 stars, last pushed 4mo ago), licensed MIT. It adds 96 tokens to every session and 729 once invoked, about $0.0005 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 2026-08-30.
Other skills, from other repositories
papr-rss
Read, search and triage the user's Papr RSS subscriptions from the shell via the papr CLI. Use when the user wants to catch up on their feeds, find or summarize articles they've subscribed to, check what's unread, star/save articles, subscribe to a new feed, or pull new posts. Triggers on: "what's in my feeds", "any…
edgeone skill scanner
A local static scanner that checks agent-skill files for security risks before they are installed or used. Static analysis examines files without running them.
dev-workflow
The complete development workflow for SkillHub contributors including local dev, staging validation, testing, and PR creation. Ensures agents follow the correct sequence of steps.
convex-performance-audit
Audits Convex performance for reads, subscriptions, write contention, and function limits. Use for slow features, insights findings, OCC conflicts, or read amplification.
video-frames
Extract a single frame from a local video at the first frame, a timestamp, or a zero-based frame index using FFmpeg.
convex-insights
Query a running Convex app's logs + health in natural language (official MCP): failures, slow/expensive functions, deploy causality — scoped, evidence-backed, with a dashboard deep link.