A command that chooses the appropriate Wheeler command for a research task. Wheeler is a research-workflow tool that can capture sources, manage investigation plans, query a knowledge graph, and help write results.

Use it to start or resume research, add papers or other evidence, discuss and execute plans, query or maintain the knowledge graph, collaborate, or write up findings.

It removes the need to know which Wheeler command fits the task. It uses the task and any saved session context to route the user through the research workflow.

wh:start — Command by maxwellsdm1867

Install

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.

agentmods

npx agentmods add commands/maxwellsdm1867/wheeler/start

Clone the repo

git clone --depth 1 https://github.com/maxwellsdm1867/wheeler

Made for: Claude Code.

Per session 19 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 563 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

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.

Model	Per session	Once invoked
Fable 5	$0.00019	$0.00563
Opus 5	$0.00010	$0.00282
Sonnet 5	$0.00004	$0.00113
Haiku 4.5	$0.00002	$0.00056

Measured 2d ago against content hash 878df0275fda, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

wh:start 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 2d 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/start.md · 40 lines

What it actually says

Wheeler Router

Route the user to the right /wh:* command. Do not ask which command to use: analyze their intent and invoke via the Skill tool.

Routing procedure

If $ARGUMENTS is non-empty, treat it as the user's task description and skip to step 3.
If $ARGUMENTS is empty:
- Check .plans/STATE.md and .wheeler/ for session context.
- Ask the user what they're working on via AskUserQuestion with 2-4 options covering the most likely intents (e.g., "starting a new investigation", "adding data to the knowledge graph", "continuing prior work", "writing up results").
Match intent to a /wh:* command using this priority:
- Session lifecycle: status/resume at session start; pause/close at session end; chat for casual discussion
- Data capture (concrete artifacts provided): add (DOI, paper, dataset, file) over note (insight, observation)
- Investigation workflow (progressive): discuss -> plan -> execute -> write
- Graph operations: ask (query), compile (synthesis), dream (maintenance), graph-link (batch orphan provenance), graph-review (quality audit)
- Collaboration: pair (interactive), handoff (background), reconvene (review)
- Meta: report (time window), triage (GitHub issues), dev-feedback (Wheeler bugs)
Invoke the chosen command via the Skill tool. Prefix with a one-line explanation of the routing choice.
Never route to queue, init, ingest, or update: those require explicit user invocation.
If the task is not Wheeler-related, say so plainly and let the user decide whether to proceed.

Style

Never use em dashes. Use colons, commas, periods, parentheses.
Be brief. The user wants to get into the right mode, not read about modes.

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.

2d ago First seen · 40 lines · 19 tokens per session scan A 878df0275fda

Subscribe to this mod's changes

wh:start is a command published in the GitHub repository maxwellsdm1867/wheeler (10 stars, last pushed 5d ago), licensed MIT. It adds 19 tokens to every session and 563 once invoked, about $0.0001 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.

Research ai-agent citations claude claude-code knowledge-graph mcp

info

Display information and statistics about a knowledge abstract.

yifanfeng97/Hyper-Extract · 0 tokens

maestro-next

Unified entry for all development intents — classify intent, assess complexity, route to the correct execution channel: /maestro-companion (lightweight), standard single run, or /maestro and /maestro-ralph (multi-step manual/orchestrated). Pure router, never runs execution loops itself.

catlog22/maestro-flow · 64 tokens

analyze-task

Parse user task description -> detect required capabilities -> build dependency graph -> design dynamic roles with role-spec metadata. Outputs structured task-analysis.json with frontmatter fields for role-spec generation.

catlog22/maestro-flow · 0 tokens

maestro-knowhow

Intent-driven knowhow precipitation — describe what you want to capture (记一个关于X的决策 / 保存这段代码模板 / 写个部署配方 / 存个调试技巧) and the workflow infers the type and records it into .workflow/knowhow/. Pure capture surface; knowhow 的管理/审计走 /maestro-knowledge；项目约束规则走 /maestro-spec add。Triggers on "knowhow capture", "知识沉淀", "沉淀经验"…

catlog22/maestro-flow · 128 tokens

implement

Direct implementation using Edit/Write/Bash tools.