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.
git clone --depth 1 https://github.com/Spark-To-Paper-Skills/paper-wikiWrote 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.
[](https://agentmods.dev/commands/spark-to-paper-skills/paper-wiki/wiki-auto)<a href="https://agentmods.dev/commands/spark-to-paper-skills/paper-wiki/wiki-auto"><img src="https://agentmods.dev/badge/commands/spark-to-paper-skills/paper-wiki/wiki-auto/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/spark-to-paper-skills/paper-wiki/wiki-auto"><img src="https://agentmods.dev/badge/commands/spark-to-paper-skills/paper-wiki/wiki-auto.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 5.1 | $0.00056 | $0.03033 |
| Opus 5 | $0.00028 | $0.01517 |
| Sonnet 5 | $0.00011 | $0.00607 |
| Haiku 4.5 | $0.00006 | $0.00303 |
Grade A, and why
wiki-auto 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 11d 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.
How it starts
The opening of the file, as written. The whole thing — 171 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Run the paper-wiki auto pipeline. This file is a Claude Code slash-command
orchestrator: it chains the existing actions end to end and defines no new
schema or judgment rule of its own — every delegated action keeps its own
contract, untrusted-source boundary, worker least-privilege, and
single-writer rules. Each delegated action runs with the explicit context
that this is an auto run (/wiki-auto 驱动), which switches on the clauses
elsewhere conditioned on the auto-run phrase; the ingestion side reuses the
instance's existing expansion_mode: auto — no new switch.
Preconditions (unmet means a clean stop, not a workaround)
- The instance is already initialized.
wiki-initstays a one-time human setup, and the Scope fence is drawn there — this command never draws or edits the fence. research.md§ Scope fence setsexpansion_mode: auto. Defaultattendedmeans stop and report;lifecycle_state: FROZENoverrides every mode — stop.- If any source needs OCR, a local or remote GPU must be available. A busy or missing GPU is a clean stop, never CPU fallback (existing invariant, unchanged).
Inbox — wiki/INBOX.md
Instance file, created on the first auto run, append-only rows
(protocol definition: auto_run.inbox in docs/llm-wiki.protocol.yaml):
| 日期 | 环节 | 对象 | 事项 | 状态 |
- Reversible decisions a human would normally make: the machine makes them
first and records one row. Fields that normally carry a human signature
are filled
auto+ date and honestly labeled（机采标注）. - Final decisions — kills, probe waivers, gold-mine/dead-end verdicts, novelty verification, lifecycle rulings — the machine makes as well: one row each plus the decision's existing ledger/fields, signature auto + date; a human retains post-hoc reversal. Only two things stay undone, as range, not authority: experiment advance (P1/P2 probes get a row only) and paper export.
- A human vetoes any machine decision afterwards by normal editing; no new mechanism.
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.
- 11d ago First seen · 171 lines · 56 tokens per session scan A 1908496621d0
wiki-auto is a command published in the GitHub repository Spark-To-Paper-Skills/paper-wiki (7 stars, last pushed 28d ago), licensed MIT. It adds 56 tokens to every session and 3,033 once invoked, about $0.0003 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
contribute
Add a thought, source, or paste. Use --project/-p to target another project without switching current. Usage: /mb:contribute [-p ] [--mode daily|concept|daily+concept].
build
Rebuild context.md for a project from sources/. Usage: /mb:build [-p ] [project-id].
init
Scaffold a new MindBase project (v2 layout). Usage: /mb:init [template] [-- mission ...].
research
Deep research on a topic — web + existing wiki + synthesis → sources/research/. Usage: /mb:research [-p ].
new-project
Create a new project with template choice — interactive flow. Usage: /mb:new-project [template].
ask
Query the MindBase wiki — cited answer using hybrid search + graph expansion. Usage: /mb:ask [-p ] [--all-projects].