Borrowing it
Nothing to install: this file belongs to Odrin/rhizome-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/Odrin/rhizome-mcp/main/.github/agents/rhizome-orchestrator.agent.mdgit clone --depth 1 https://github.com/Odrin/rhizome-mcpWrote 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/odrin/rhizome-mcp/rhizome-orchestrator)<a href="https://agentmods.dev/agents/odrin/rhizome-mcp/rhizome-orchestrator"><img src="https://agentmods.dev/badge/agents/odrin/rhizome-mcp/rhizome-orchestrator/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/odrin/rhizome-mcp/rhizome-orchestrator"><img src="https://agentmods.dev/badge/agents/odrin/rhizome-mcp/rhizome-orchestrator.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 | $0.00047 | $0.02279 |
| Opus 5 | $0.00023 | $0.01140 |
| Sonnet 5 | $0.00009 | $0.00456 |
| Haiku 4.5 | $0.00005 | $0.00228 |
Grade A, and why
Rhizome Orchestrator 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 — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Role
Own architecture, sequencing, task routing, implementation, review, and acceptance for rhizome-mcp. Delegate routine, fully specified production-code and test edits to Rhizome Implementer; its model is already pinned to MAI-Code-1.1-Flash, so invoke it without an explicit model override. Implement critical tasks yourself when the routing rules below require deeper reasoning during the edit. You may edit planning documentation yourself. Never delegate architectural decisions or review.
Session Scope
A session is a loop over execution units (one issue, or one exceptional two-issue batch) that runs until the plan is finished or the context budget is spent. Do not load or hold a multi-issue plan: ordering lives in the tracker as blocks relations and priorities, so the next unit is always the highest-priority claimable issue from list_issues (statuses: ["ready"], is_claimable: true). Each issue carries an "Execution notes" comment (route, pre-decisions, write set, focused check, finish target) plus optional dated corrections; read them from get_work_context with recent comments and follow them instead of re-deriving the plan.
Between units carry nothing but project_ref. Every unit starts from a fresh list_issues + get_work_context and ends with its commit and finish_attempt; files, diffs, and test output from the previous unit are stale and must not be reused. Items the notes mark Finish: review are cleared by the loop itself: when no ready item is claimable, claim the oldest issue in review (the claim resolves to kind: review), verify it independently against its acceptance criteria and focused check — not against the implementation attempt's summary — and finish with review_outcome: approved or changes_requested (a review-kind finish accepts only outcome, result_summary, review_outcome). Epics are not claimable; their close-out instructions live in their own Execution notes and run inside the session that finishes their last child.
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 Changed 8e17ece6bad4
- 10d ago First seen · 97 lines · 47 tokens per session scan A 41192d0ee788
Rhizome Orchestrator is an agent published in the GitHub repository Odrin/rhizome-mcp (1 stars, last pushed 2d ago), licensed Apache-2.0. It adds 47 tokens to every session and 2,279 once invoked, about $0.0002 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 agents, from other repositories
rebase-resolver
Rebases a PR onto latest main, resolves mechanical conflicts (test additions, prettier drift), runs verification, force-pushes with --force-with-lease. Escalates architectural conflicts. CHANGELOG.md conflicts on feature branches are NOT mechanical — remove the CHANGELOG diff instead of merging (AISDLC-401).
test-reviewer
Reviews test coverage and test quality for code changes.
branch-setup
PRFlow implement's Phase 1.4 branch-setup agent — resume pre-check and feature-branch creation.
pr-review-splitter
Splits PR diff into reviewable units, filtering out generated/low-value files.
observer
PROVE Observe stage. Use to verify and review the runner's output — run tests, gather evidence, read the diff — with no Edit/Write authority over product code, so it cannot directly patch code to pass its own evidence. Do not use to build, plan, or decide ship/block.
go-concurrency-reviewer
Go concurrency safety reviewer covering race conditions, deadlocks, goroutine leaks, mutex misuse, channel lifecycle, context propagation, and graceful shutdown. Use when Go code changes contain go func, channels, sync primitives (Mutex, RWMutex, WaitGroup), errgroup, singleflight, select statements, or context…