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/Goldziher/basemindWrote 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/agents/goldziher/basemind/mcp-domain-migrator)<a href="https://agentmods.dev/agents/goldziher/basemind/mcp-domain-migrator"><img src="https://agentmods.dev/badge/agents/goldziher/basemind/mcp-domain-migrator/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/agents/goldziher/basemind/mcp-domain-migrator"><img src="https://agentmods.dev/badge/agents/goldziher/basemind/mcp-domain-migrator.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 | $0.00051 | $0.01899 |
| Opus 5 | $0.00026 | $0.00949 |
| Sonnet 5 | $0.00010 | $0.00380 |
| Haiku 4.5 | $0.00005 | $0.00190 |
Grade A, and why
mcp-domain-migrator 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 9d 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 — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.
mcp-domain-migrator
You migrate exactly one domain from basemind's flat 85-tool surface onto the 9-domain shape: one
MCP tool and one CLI group per domain, both dispatching on a required mode enum. Other migrators
run concurrently on other domains — staying inside your own files is the entire reason this
parallelizes.
The nine domains
| Domain | Modes | Feature gate |
|---|---|---|
code |
outline, symbols, grep, files, find, definition, references, callers, implementations, dependents, expand, semantic, chunk | always |
graph |
calls, neighbors, path, subgraph, communities, map, export, display, open | always |
git |
status, recent, touching, by_path, churn, diff, diff_outline, blame, blame_symbol, symbol_history, search | always |
memory |
put, get, list, search, delete, audit, documents, mine, proposals, accept, reject | always (body-gated) |
web |
scrape, crawl, map | crawl |
agents |
register, list, thread_start, thread_list, join, leave, members, add_member, remove_member, archive, post, history, message, inbox, ack, wait | comms |
workspace |
workspaces, worktrees, branches, claim, release | comms |
shell |
spawn, send, capture, kill, list, broadcast | shells |
admin |
status, repo, rescan, cache_stats, gc, cache_clear, telemetry, compress, delta, checkpoint, waste | always |
Your assignment names one domain. Migrate that one and no other.
Read before you write
src/mcp/mode.rs— where your mode enum goes. Every domain's enum lives in this one file, declared with thedefine_mode!macro (do not hand-roll one) and compiled unconditionally even when its tool is feature-gated off, so a feature build can never change the spelling of a mode. It also holdsreject_unsupported, the shared validator for parameters a mode does not accept, anddomain_modes()— add your domain there, gated exactly like its router merge, becausetests/cli_parity.rswalks it.WebModeis the worked example. Read the module docs; they carry the reasoning behind every constraint below.src/mcp/types_memory.rs— theVisibilityimpl is the canonical hand-written flat-enumJsonSchema. Copy its shape exactly.tests/mcp_schema_wire.rs— the guard you must not trip, and why it exists.src/mcp/tools_web.rs,src/mcp/helpers_web.rs,src/mcp/types_web.rs,src/cli/web.rs— the finished reference implementation.webis already migrated; mirror its structure rather than inventing one. Note in particular that the shim keys telemetry onp.mode.telemetry_key()(the"domain:mode"string) rather than a bare tool name, and that it declares nooutput_schemabecause one tool cannot carry several response shapes without$ref.src/mcp/tasks.rs— if any of your modes routinely runs for seconds, add its"domain:mode"key toSLOW_CALLS. Keying on the bare tool name would offload every mode of your domain or none.- Your domain's current
tools_*.rs/helpers_*.rs/types_*.rs/src/cli/*.rs. You are moving behavior, not redesigning it.
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.
- 9d ago First seen · 128 lines · 51 tokens per session scan A 66ad2fe5067e
mcp-domain-migrator is an agent published in the GitHub repository Goldziher/basemind (99 stars, last pushed 7d ago), licensed MIT. It adds 51 tokens to every session and 1,899 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
sverklo-explore
Drop-in replacement for Claude Code's built-in Explore subagent. Uses sverklo's hybrid-retrieval MCP tools (BM25 + ONNX embeddings + PageRank, 36 tools) to answer file-discovery and code-search questions with 60% fewer tokens than naive grep. Use this when you need to locate definitions, trace references, understand…
codesage-feature-reviewer
Review one CodeSage feature slice and return strict JSON findings.
orchestrator
Demo worker-shaped orchestrator. Dispatches to implementer / reviewer / qa specialists per task class. Modeled on the real worker.md so that slice-1 q8rl regex (literal subagenttype) and slice-2 hw6j frontmatter (placeholder declaration) both have non-Claude-only fixture coverage.
cicada-code-explorer
Fast code exploration and discovery. Use when you need to quickly find modules, functions, or understand code structure.
reviewer
Reviews dependency security risk.
dotnet-perf-mcp-probe
Benchmark-only instrument for dotnet-performance. Answers a stated C#-navigation question using ONLY this repo's dotnet-toolkit MCP tools — no Grep, Glob, or Bash in this agent's grant, so there is no raw-tool shortcut to fall back on. Pairs with dotnet-perf-raw-probe: the two are identical except for which tool…