daily-digest

<!-- autoresearch: variation B — curatorial discipline (filter → distill → structure → sanity-check) folded with sandbox-safe inputs and memory-aware dedup --> > **${var}** — Topic for the digest (e.g. "AI agents", "solana", "rust"). **Required** — set your topic in aeon.yml. Today is ${today}. Generate and send a daily **${var}** digest. The whole point of a digest is **signal, not volume**. A reader skimming for 60 seconds should walk away with three things they didn't know that morning and one of them should change a decision they'd make this week. Anything that doesn't clear that bar gets cut. ## Phase 1 — Gather (cast a wide net) Pull from at least two of these source classes — never rely on a single one: 1. **WebSearch** (built-in) — run 2 distinct queries: - `"${var}" news ${today}` (broad) - One narrower query you choose based on `${var}` (e.g. for "solana" → `"solana" launches OR funding OR exploit ${today}`; for "AI agents" → `"agent framework" OR "agentic" release ${today}`) 2. **xAI x_search via Grok** — pulls the X/Twitter signal layer. - **Preferred path (sandbox-safe):** read `.xai-cache/digest.json` if it exists. The workflow's `scripts/prefetch-xai.sh` populates it before Claude runs. If you find the cache empty or absent, log a one-line note and continue — do not retry curl in a loop. - **Fallback:** if the cache is missing, attempt a WebFetch to a public X search URL like `https://x.com/search?q=${var}&f=live` and extract a few top posts. ...