agentsclimarketplace
Sign in

Miller search debug

Skill anortham/miller/skills/miller-search-debug

Use when Miller search misses expected results, ranks surprising hits, returns noisy output, or needs mode/content/region troubleshooting.From its SKILL.md

Install
npx -y skills add anortham/miller --skill miller-search-debug

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 3 stars3 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.7 KB, as published. Nobody here has run it

Miller Search Debug

Diagnose search behavior by checking the query mode, index freshness, projection limits, and whether the expected result exists.

Workflow

  1. Reproduce the exact query:
search(query="<original>", limit=20)
  1. Compare modes when the query intent is unclear:
search(query="<symbol-ish>", mode="symbol")
search(query="<path-ish>", mode="file")
search(query="<docs or prose>", mode="content")
search(query="<source body text>", mode="source")
search(query="<imported log text>", mode="external")
search(query="<imported web text>", mode="web")
search(query="<broad text>", mode="all-text")
search(query="<comment or literal>", regions="comment|string_literal|doc_comment")
search(query="<known area>", file_pattern="src/ui/**", language="typescript")
  1. Check common beta gotchas:
  • Natural-language search hides test code by default; use exclude_tests=false when tests are expected.
  • Symbol search ranks name + signature; docs/prose belong in mode=content.
  • Source bodies belong in mode=source; imported logs/reports and web markdown belong in mode=external or mode=web.
  • Cross-workspace exact-text audits should use content search "<term>" --workspace-id all --kind source|docs|config|external_file|web.
  • Comment, doc-comment, and string-literal searches require region indexing and a fresh sidecar.
  • File/path queries should use mode=file when auto mode looks noisy.
  • Scoped workflows should use file_pattern and language before raising limit.
  • Stale indexes should be refreshed with workspace(operation="refresh").
  1. Verify the expected item exists:
search(query="<exact symbol or file>", mode="auto", exclude_tests=false)
inspect(target="<expected-symbol-or-file>")
  1. If the expected result is connected by calls rather than text, use:
trace(target="<nearby-symbol>")
context(query="<workflow around expected result>")

Report

Use this shape:

Search Debug: "<query>"
Expected: <symbol/file/text>

Observed:
- <top results or failure>

Diagnosis:
- <wrong mode, stale index, hidden tests, projection limit, region disabled, or true miss>

Next:
- <specific query, refresh, or implementation follow-up>

Do not widen Miller's symbol ranking as a first response. First prove that mode=content, regions=..., file mode, or refresh does not solve the workflow.

Keep looking

Skills are one crate of 627,801. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.