← All examples Levirge Search · examples

The nightly release digest

A production run from question to posted digest: two selected reads (28 KB of a 1.26 MB response), the constructed URL ordinary tools refuse, the generated summary with its sources, and the measured cost.

1

The research question

A scheduled job wakes nightly and asks: "What changed in llama.cpp since the tag we digested last night?" — then posts a summary the team actually reads. Captured from a production run, re-verified 8 Aug 2026; payloads verbatim, long bodies excerpted.

2

The ordinary tool — refused before it starts

The job must construct compare/<last>...<tonight> from two tags. Native fetch tools only accept URLs that appeared verbatim in the conversation — a constructed URL is refused outright, and even when accepted, the body comes back empty with nothing to parse.

native web_fetch refused
// same URL, host's built-in fetch
web_fetch("https://api.github.com/repos/ggml-org/llama.cpp/compare/b10256...b10308")

✗ URL not present in conversation — refused
✗ retried from a search hit: empty body, no JSON
3

Call one — tonight's tag: three fields, not thirty

The release object carries 30+ fields — assets with download counts, full release notes, author objects. The job needs three. select trims server-side, so only these reach the agent's context.

without select ~30 KB · full object
// full response, truncated
fetch("https://api.github.com/repos/ggml-org/llama.cpp/releases/latest")

{ "url": "https://api.github.com/repos/…",
  "author": { "login": "…", … 16 more fields },
  "tag_name": "b10308",
  "name": "b10308", "draft": false,
  "published_at": "2026-08-07T14:05:22Z",
  "assets": [
    { "name": "llama-b10308-bin-macos-arm64.zip",
      "download_count": …, … per-asset fields },
    … more assets ],
  "body": "…full release notes…" }
with select 3 fields
// select trims server-side
fetch("https://api.github.com/repos/ggml-org/llama.cpp/releases/latest",
      select: ["tag_name", "name", "published_at"])

{ "tag_name": "b10308",
  "name": "b10308",
  "published_at": "2026-08-07T14:05:22Z" }
4

Call two — the diff, without the megabyte

The compare object is 52 commits, each with author objects, trees and verification blocks, then 183 per-file patches. The jq-lite path commits[].commit.message reaches into every array element and preserves the nested shape, so rows stay aligned. Same request, side by side:

without select 1.26 MB · ~315k tokens
// the same request, no select — truncated
fetch("https://api.github.com/repos/ggml-org/llama.cpp/compare/b10256...b10308")

{ "url": "…", "html_url": "…",
  "base_commit": { "sha": "…",
    "commit": { "author": {…}, "committer": {…},
      "tree": {…}, "verification": {…} }, … },
  "status": "ahead", "ahead_by": 52,
  "commits": [
    { "sha": "22dc605c4ead…", "node_id": "…",
      "commit": {…}, "author": {…},
      "committer": {…}, "parents": […] },
    … 51 more ],
  "files": [
    { "filename": ".github/workflows/build-apple.yml",
      "patch": "@@ -541,7 +541,9 @@ …" },
    … 182 more ] }
with select 28 KB · ~7k tokens
// the same request, with select
fetch("https://api.github.com/repos/ggml-org/llama.cpp/compare/b10256...b10308",
      select: ["ahead_by", "total_commits",
               "commits[].sha", "commits[].commit.message"])

{ "ahead_by": 52,
  "total_commits": 52,
  "commits": [
    { "sha": "22dc605c4ead…",
      "commit": { "message":
        "ci: fix vulkan llvmpipe runs (#26533)" } },
    { "sha": "935cad6497e8…",
      "commit": { "message":
        "llama : move n_vocab from llama_sampler…" } },
    … 50 more ] }
45× smaller — the trim happens server-side; the megabyte never reaches the agent.
5

The digest the job posts

Generated from the 52 captured commit messages — every line traces to a PR number in the payload above.

posted digest cited
llama.cpp nightly — b10256 → b10308 · 52 commits

Breaking:  Qwen3-TTS lands in mtmd; the llama-tts
           binary changes (#26254)
Backends:  Vulkan GATED_LINEAR_ATTN (#25601); SYCL DSv4
           ops (#26568) + set_rows types (#26515);
           AMD ROCm gfx1151 joins CI (#26544)
Hardening: gguf-py validates n_dims, guards uint64
           overflow on crafted files (#25401); server
           file-search walk hardened (#26626); UI dep
           bumps clear vulnerable packages (#26607)
Server/UI: spec-decode counters on /metrics (#26389);
           per-conversation working directory (#26518)

Sources: releases/latest + compare/b10256...b10308,
read via select — 28 KB of 1.26 MB reached the agent.
6

Measured cost, and where this came from

Per run: two selected reads ≈ 28 KB / ~7k tokens into context, against ~1.29 MB / ~320k tokens without select — and the compare call wouldn't have happened at all on a native fetch tool.

provenance re-verified 8 Aug 2026
origin      captured from a production run
re-verified 8 Aug 2026, Claude Code over MCP:
            schema: true → commits[] $items: 52 · files[] $items: 183
            select rerun → identical ahead_by: 52 payload
retention   bodies cached server-side · 24 h