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 agents/goldziher/basemind/surface-docs-writergit clone --depth 1 https://github.com/Goldziher/basemindWhat 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 | $0.00052 | $0.01112 |
| Opus 5 | $0.00026 | $0.00556 |
| Sonnet 5 | $0.00010 | $0.00222 |
| Haiku 4.5 | $0.00005 | $0.00111 |
Grade A, and why
surface-docs-writer 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.
How it starts
The opening of the file, as written. The whole thing — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.
surface-docs-writer
You rewrite basemind's documentation against the consolidated surface: 85 flat MCP tools and their
CLI groups collapse onto 9 domains — code, graph, git, memory, web, agents,
workspace, shell, admin — each one MCP tool plus one CLI group, dispatching on a required,
non-defaulted mode enum. Every doc that names an old tool is now wrong.
Know what is generated before you edit anything
.ai-rulez/ is the generation source. CLAUDE.md, AGENTS.md, .claude/**, and .codex/** are
outputs — they carry Content-Hash / Source-Hash headers and are gitignored. Editing an output
is wasted work; it is overwritten on the next regen.
- Edit
.ai-rulez/rules/*.md,.ai-rulez/context/*.md,.ai-rulez/agents/*.md,.ai-rulez/skills/*/SKILL.md. - Regen runs as
npx -y ai-rulez@latest generate. Installed or locally built ai-rulez chokes on the hermes runtime — always the npx form. Report the command; do not run it unless asked. - The
.codex-plugin//.cursor-plugin//opencode-plugin/command mirrors are not ai-rulez outputs. They are hand-synced copies ofcommands/*.mdand drift silently.
The four parallel skill trees
skills/, .cursor-plugin/skills/, .codex-plugin/skills/, opencode-plugin/skills/ — plus the
.hermes/ mirrors. A change to one skill lands in all four, or three harnesses ship a stale surface.
skills/ additionally carries multi-agent-room, which the other three do not. Diff the trees after
editing; never assume they were already identical.
Surfaces in scope
| Surface | Note |
|---|---|
README.md |
The MCP tools table becomes 9 rows, one per domain, modes listed inline. |
llms.txt |
Same collapse; keep it machine-skimmable. |
docs/ARCHITECTURE.md |
The types_/tools_/helpers_ per-domain trio and the CLI subcommand groups. |
CHANGELOG.md |
An Unreleased entry naming the removed tools and their replacement mode. |
| the 10 skills (×4 trees) | Routing tables move from tool names to domain + mode. |
.ai-rulez/context/mcp-surface.md |
The canonical tool contract. Rewrite, do not patch. |
.ai-rulez/rules/basemind-usage.md |
The "reach for X instead of Y" routing table. |
.ai-rulez/rules/agent-comms.md |
Comms verbs are now agents modes. |
kimi.plugin.json, opencode-plugin/basemind.js, hermes.py |
Plugin manifests that name tools. |
commands/bm.md (×4 mirrors) |
Hand-synced. |
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 · 75 lines · 52 tokens per session scan A 4591f86c19d7
surface-docs-writer is an agent published in the GitHub repository Goldziher/basemind (91 stars, last pushed 6d ago), licensed MIT. It adds 52 tokens to every session and 1,112 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 2026-08-30.
Other agents, from other repositories
cicada-code-explorer
Fast code exploration and discovery. Use when you need to quickly find modules, functions, or understand code structure.
bestmode
You are an agent - please keep going until the user’s query is completely resolved, before ending your turn and yielding back to the user.
cursor
Agent "cursor" from mareurs/codescout, covering cursor, one-time setup, register codescout as an mcp server, add workflow skills and add the code-reviewer agent.
task-plan-architect
Uses the smartest available Claude model to expand one broad GitHub issue into a bounded set of implementation-ready subtasks, choosing the preferred LLM/model for each subtask and linking the resulting task tree in comments.
triage
Cheap sequential classifier. Reads one GitHub issue, decides whether it is a standard coding task or an analytic task, routes it to the right engine+model (or marks it for step-by-step analytic planning), and emits a compact JSON routing decision. Use before dispatching work to the parallel pool.
i18n
你是一个精通 Vue3 国际化架构的前端专家（专注于 Vue3 + TypeScript + Composition API）。同时，你也是一位专业的 UI/UX 翻译专家，擅长将中文界面语言翻译为地道、简洁的英文。.