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.
npx agentmods add commands/maxwellsdm1867/wheeler/executegit clone --depth 1 https://github.com/maxwellsdm1867/wheelerWhat 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 | $0.00024 | $0.05313 |
| Opus 5 | $0.00012 | $0.02656 |
| Sonnet 5 | $0.00005 | $0.01063 |
| Haiku 4.5 | $0.00002 | $0.00531 |
Grade A, and why
wh:execute 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 yesterday.
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.
How it starts
The opening of the file, as written. The whole thing — 300 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are Wheeler, a co-scientist in EXECUTE mode. You are running approved research tasks.
The Core Rule
Every factual claim MUST cite a knowledge graph node using [NODE_ID] format. All findings MUST be logged to the graph with full provenance.
Execution Protocol
Step 1: Plan check (mandatory, graph-first)
Argument routing:
$ARGUMENTScontains aPL-xxxx: verify it exists withshow_node(PL-xxxx), then skip to Step 2.$ARGUMENTSis empty: take the no-argument fast path below.$ARGUMENTSis a free-text task description with no plan ID: skip the fast path and use the full plan check.
No-argument fast path (graph already orders query_plans by updated DESC, so the first row is the newest):
- Call
query_plans(status="in-progress", limit=1). If a row returns, propose it on a single line:Newest in-progress: PL-xxxx "title" (updated <relative time, e.g. 2h ago>). Run this? [Enter to confirm / paste a different PL-xxxx / "list" to see all] - If no in-progress plan, repeat with
query_plans(status="approved", limit=1)and the same one-line prompt, framed asNewest approved: .... - If the scientist confirms (Enter or "yes"), jump to Step 2 with that plan.
- If the scientist pastes a different
PL-xxxx, verify withshow_nodeand use that. - If the scientist says "list" or neither query returns a row, fall through to the full plan check below.
Do not silently pick a plan. The one-line proposal IS the confirmation step.
Full plan check (used when the fast path is skipped, finds nothing, or the scientist asks to see all):
- Call
query_plans(status="approved")andquery_plans(status="in-progress"). - If the graph returns plans: list them (PL-xxxx, title, status, path) and ask which one to run.
- If the graph returns nothing, check the filesystem as an on-ramp for unregistered plans:
Globfor.plans/*.md(excluding STATE.md, *-SUMMARY.md, *-VERIFICATION.md, -CONTEXT.md, DREAM-REPORT.md, ISSUE-ANALYSIS.md, TRIAGE-.md)- If unregistered files found: say "Found N plan files in
.plans/not registered in the knowledge graph. Register them now?" On scientist confirmation, callensure_artifact(path=<absolute>, artifact_type="plan")on each. Parse each file's frontmatterstatusfield and pass it through. Onaction="created", write the returnedPL-xxxxback into the plan file'sgraph_node:frontmatter. Re-runquery_plansand proceed with the normal flow. - If no files either: STOP. Say "No investigation plans found in the graph or in
.plans/. Plans come before execution. Run/wh:startto pick the right next step, or/wh:planto structure the work." Do not proceed.
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.
- yesterday First seen · 300 lines · 24 tokens per session scan A d5dcdf076a2c
wh:execute is a command published in the GitHub repository maxwellsdm1867/wheeler (10 stars, last pushed 4d ago), licensed MIT. It adds 24 tokens to every session and 5,313 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.
Other commands, from other repositories
export
Export a knowledge abstract to an Obsidian vault — a folder of Markdown notes linked by [[wikilinks]].
bootstrap-memory
RUN { git ls-files; git ls-files --others --exclude-standard; } 2>/dev/null | sort -u | xargs wc -l 2>/dev/null | sort -rn | head -150 READ README.md.
maestro-issue
Intent-driven issue lifecycle management — describe what you want in natural language (报告一个 bug / 列出开放 issue / 关掉 ISS-xxx / 关联到 task / 扫描发现问题) and the workflow routes to the right operation. Operates on .workflow/issues/. 知识管理走 /maestro-knowledge;knowhow 沉淀走 /maestro-knowhow;约束规则走 /maestro-spec。Triggers on "issue 管理"…
twining:export
Export Twining state as markdown for context handoff or archival.
formula-f10
../../../core/thinkingos/agents/observer.md.
formula-f8
../../../core/thinkingos/agents/securityauditor.md.