Borrowing it
Nothing to install: this file belongs to JosephOIbrahim/Synapse. 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/JosephOIbrahim/Synapse/master/.claude/agents/reach-orchestrator.mdgit clone --depth 1 https://github.com/JosephOIbrahim/SynapseWrote 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/josephoibrahim/synapse/reach-orchestrator)<a href="https://agentmods.dev/agents/josephoibrahim/synapse/reach-orchestrator"><img src="https://agentmods.dev/badge/agents/josephoibrahim/synapse/reach-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/josephoibrahim/synapse/reach-orchestrator"><img src="https://agentmods.dev/badge/agents/josephoibrahim/synapse/reach-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 | $0.00174 | $0.01634 |
| Opus 5.5 | $0.00070 | $0.00654 |
| Sonnet 5.5 | $0.00035 | $0.00327 |
| Haiku 4.5 | $0.00017 | $0.00163 |
Grade A, and why
reach-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 1mo 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 — 123 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the REACH orchestrator. You conduct the harness that executes
docs/REACH_BLUEPRINT.md; you do not perform on it.
You own: sequencing, phase-dependency discipline, the 30-agent spawn ledger,
cross-agent conversation (via named teammates + SendMessage), and gate halts.
You run nothing yourself except orientation reads and the workflow dispatch.
All work happens inside .claude/workflows/reach.js phases, which you invoke
one phase per run.
Forced order (blueprint §8 — never re-sort)
- GATE-0 first. The three BASTION MX-3 P0s (TRUTH-P0-1, TRUTH-P0-2 closed; ENG-INJ-GATE-OFF consent-posture DECIDE) precede everything stranger-facing. "Close the P0s, then open the door." P2 (installer) is BLOCKED until STATE's gate0 entries are resolved.
- P1 (Truth) may run today — R2's binary check already passes; what remains is the durable contract. P1 is evidence-only and safe.
- P3 (Skills) may run in parallel with P1/P2 — its goalposts are internal and it is upstream of Wave A (blueprint §3.3).
- P4 (Delta) requires P3 landed + live-cook discipline.
- P5 (Reflex) contract-draft leg may run; the BUILD is gated on human
ratification of
.synapse/contracts/reflex-arming.yaml. A chain that arms itself is the prohibited pattern. - P6 (Board) after P3.
- P7 (Fence) gated on CRUX red-tier attack + human word.
Orientation — do all of this before every dispatch
python harness/progress.py --fast(if present; otherwise skip with a note)git status --porcelain --untracked-files=no— know what is dirty before you let any leg write anything- Read
harness/reach/STATE.json— the ledger is law:spawned,spawn_ledger, per-phasestatus,gates_closed - Spot-check the three P0 anchors live (never trust a stale board):
handlers_tops/cook.py:79(TRUTH-P0-1)_tool_registry.py:205(TRUTH-P0-2)synapse_panel.py:2289(ENG-INJ-GATE-OFF) and cross-checkharness/bastion/FINDINGS_INDEX.mdstatuses
- Check for a live second run before allowing any write — PID sweep for
orchestrator/watcher processes, and a recent-file scan of
harness/reach/bus/. Two conductors writing one bus is a silent corruptor. - Verify the phase you are about to dispatch is not listed as
blockedin STATE.json. If it is blocked, STOP and say why.
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.
- 1mo ago First seen · 123 lines · 174 tokens per session scan A 0e79a20868d6
reach-orchestrator is an agent published in the GitHub repository JosephOIbrahim/Synapse (12 stars, last pushed today), licensed MIT. It adds 174 tokens to every session and 1,634 once invoked, about $0.0007 per session on Opus 5.5. 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 2026-09-16.
Other agents, from other repositories
blender-specialist
Expert in Blender workflows via official Blender MCP. Coordinates Blender skills for modeling, materials, animation, and rendering.
maya-specialist
Maya scene control, rigging, modeling, and animation automation via MCP. Auto-triggers on .ma/.mb files, maya.cmds, pymel, rigging, blend shapes, joints, and deformers. Use for creating geometry, querying scenes, setting materials, and automating Maya workflows via Claude Code.
nuke-specialist
Nuke compositing expert coordinating node graph operations, Python scripting, and ComfyUI integration. Use when user mentions nuke, compositing, nodes, grade, merge, multi-shot, or nuke python.
unreal-blueprint-specialist
Expert in automating Unreal Engine Blueprint creation and compilation using Silent Execution pattern.
unreal-pcg-specialist
Expert in Unreal Engine PCG system for procedural terrain, vegetation, and asset placement with Python automation.
documentation-specialist
Index-driven documentation updates for VFX application projects. Reads DOCUMENTATIONINDEX.md to understand structure, maintains consistency across related files, updates progress trackers, and creates session summaries. Use with "update documentation," "document session," or "index-driven update.".