wheeler: Command for Claude Code

.claude/commands/wh/asta-scholar.md

wh:asta-scholar is a command for Claude Code from maxwellsdm1867/wheeler. It costs 43 tokens per session (1,927 once invoked), scanned A, original, MIT.

A Wheeler command for looking up scientific papers through Asta and Semantic Scholar, a research-paper search service, then recording the results in Wheeler's knowledge graph.

In plain words
What is it for?
Use it to retrieve one paper, search for papers, find papers that cite a paper, search for matching passages, or list an author's papers.
Why use it?
It keeps paper searches, citation links, author results, and matching passages connected to the investigation's existing context.

Command for Claude Code

Written for Claude Code: allowed-tools in frontmatter.

This is maxwellsdm1867/wheeler's own configuration. It tells Claude Code how to work on wheeler itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything wheeler configures →

Part of the wh plugin — 41 skills, 40 commands, 2 agents, 3 hooks, 5 MCP servers shipped together

Reuse

Borrowing it

Nothing to install: this file belongs to maxwellsdm1867/wheeler. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/maxwellsdm1867/wheeler/main/.claude/commands/wh/asta-scholar.md
Clone the repo
git clone --depth 1 https://github.com/maxwellsdm1867/wheeler

Made for: Claude Code.

Or install wh, the plugin that ships this one along with the rest of its 41 skills, 40 commands, 2 agents, 3 hooks, 5 MCP servers.

Wrote 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.

agentmods badge for wh:asta-scholar

README.md
[![agentmods](https://agentmods.dev/badge/commands/maxwellsdm1867/wheeler/asta-scholar.svg)](https://agentmods.dev/commands/maxwellsdm1867/wheeler/asta-scholar)
Your own site
<a href="https://agentmods.dev/commands/maxwellsdm1867/wheeler/asta-scholar"><img src="https://agentmods.dev/badge/commands/maxwellsdm1867/wheeler/asta-scholar.svg" alt="Measured on agentmods" height="20"></a>
Per session 43 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,927 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00043 $0.01927
Opus 5 $0.00022 $0.00963
Sonnet 5 $0.00009 $0.00385
Haiku 4.5 $0.00004 $0.00193

Measured 8d ago against content hash ac1aba649aaa, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

wh:asta-scholar 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 8d 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.

.claude/commands/wh/asta-scholar.md · 99 lines

How it starts

The opening of the file, as written. The whole thing — 99 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are Wheeler, querying Semantic Scholar through the Asta CLI and marshalling the results into the knowledge graph. You orchestrate; the asta CLI calls Semantic Scholar and owns its own auth and timeouts; one deterministic wheeler integrate verb writes the graph.

Semantic Scholar has five sub-queries. The ingest auto-detects which one from the artifact shape, so you only pick the right CLI call:

  • get: one paper by id or DOI.
  • search: a relevance-ranked paper list for a query.
  • citations: the papers that cite a target paper (builds the citation graph).
  • snippet: passage-level matches for a query (each becomes a Finding linked to its paper).
  • author: an author's papers by author id (each becomes a Paper; the queried author is recorded on the run).

Preflight

  1. Confirm Asta is installed: asta --version. If that fails, say Asta is not available and stop. Do not attempt the query.
  2. Read context so the query is informed by the graph. Use mcp__wheeler_core__search_context with the user's topic (or the active question) to see what is already known, and mcp__wheeler_query__query_papers to see which papers are already recorded. Use this only to sharpen the query and to choose a link target; do not invent results.

Always request corpusId

corpusId is NOT in the Semantic Scholar default field set: it only appears when you ask for it. Dedupe across Paper Finder, Theorizer, and Semantic Scholar keys on corpus_id, so you MUST request it in every call, for example --fields corpusId,title,authors,year,venue,citationCount. Without it, papers will not match across services. snippet-search already returns corpusId, but keep the field list explicit for the others.

Choose the sub-query, run it, and ingest

Pick at most one link target: the Question (Q-...) or Plan (PL-...) this query supports. Each relevant result links RELEVANT_TO that node. Omit --link-to if there is no clear target. Pass --used with the graph node ids the query was built from (at minimum the link target, the Q-/PL- that motivated it): this records Execution -[USED]-> each input (input-side provenance), so every result traces back to the graph context that shaped the query. Omit --used if there were no graph inputs. The verb is idempotent: re-running the same artifact creates no duplicate papers, findings, edges, or USED edges (papers dedupe on corpus_id, snippet findings on a content hash, edges are guarded by link_once).

Read the full file on GitHub · 99 lines

Changes

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.

  1. 8d ago First seen · 99 lines · 43 tokens per session scan A ac1aba649aaa

Subscribe to this mod's changes

wh:asta-scholar is a command published in the GitHub repository maxwellsdm1867/wheeler (11 stars, last pushed 6d ago), licensed MIT. It adds 43 tokens to every session and 1,927 once invoked, about $0.0002 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 2026-08-31.