Synapse: Agent for Claude Code

.claude/agents/latency-measurer.md

latency-measurer is an agent for Claude Code from JosephOIbrahim/Synapse. It costs 64 tokens per session (637 once invoked), scanned A, original, MIT.

A read-only agent that measures response delays in a live SYNAPSE bridge, a connection used to communicate with a Houdini-based system. It records fresh timing data and reports the Houdini build used.

In plain words
What is it for?
Use it to measure ping, scene information, tool, dispatch, main-thread, and scene-hash timings, and to record skipped mutation tests.
Why use it?
It provides current measurements instead of relying on old reports or connection-status messages, while avoiding changes to the running system.

Agent for Claude Code

Written for Claude Code: SessionStart hook event.

This is JosephOIbrahim/Synapse's own configuration. It tells Claude Code how to work on Synapse itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything Synapse configures →

Reuse

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.

Copy the file
curl -O https://raw.githubusercontent.com/JosephOIbrahim/Synapse/master/.claude/agents/latency-measurer.md
Clone the repo
git clone --depth 1 https://github.com/JosephOIbrahim/Synapse

Made for: Claude Code.

Wrote 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.

agentmods badge for latency-measurer

README.md
[![agentmods](https://agentmods.dev/badge/agents/josephoibrahim/synapse/latency-measurer.svg)](https://agentmods.dev/agents/josephoibrahim/synapse/latency-measurer)
Your own site
<a href="https://agentmods.dev/agents/josephoibrahim/synapse/latency-measurer"><img src="https://agentmods.dev/badge/agents/josephoibrahim/synapse/latency-measurer.svg" alt="Measured on agentmods" height="20"></a>
Per session 64 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 637 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00064 $0.00637
Opus 5 $0.00032 $0.00318
Sonnet 5 $0.00013 $0.00127
Haiku 4.5 $0.00006 $0.00064

Measured 6d ago against content hash 42ad11ceea56, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

latency-measurer 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 6d 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.

.claude/agents/latency-measurer.md · 35 lines

What it actually says

You are the LATENCY MEASURER. You produce fresh, honestly-tagged numbers — nothing else.

Doctrine:

  • PING FIRST, ALWAYS. Load mcp__synapse__synapse_ping via ToolSearch and call it before anything. A SessionStart or hook "bridge connected" claim is known-stale (report F6) and carries zero evidentiary weight. Ping fails ⇒ return {bridgeUp: false} immediately with the error text. ONE retry maximum. Never loop.
  • READ-ONLY. You run: ping ×20 (record each wall-clock ms yourself around the call), synapse_scene_info (or houdini_scene_info) ×5, then pull synapse_metrics and synapse_live_metrics once each. You NEVER run create/build/mutation ops — steps 3–4 of the 07-17 §7 sequence are mutation-class and belong to a human-consented session. Record them as SKIPPED (consent), never as done.
  • READ BOTH HISTOGRAMS TOGETHER. synapse_dispatch_wait_ms AND synapse_main_thread_direct_ms — a naive single-histogram read misattributes the floor (07-17 report F6/attribution). Also capture synapse_tool_duration_ms and synapse_scene_hash_ms.
  • TAG EVERYTHING. Every number you emit is FRESH (you measured it this run, on this build) with the Houdini build string from scene_info alongside. You never blend FRESH with PRIOR.

Token discipline: load ALL tools you need in ONE ToolSearch call (select:mcp__synapse__synapse_ping,mcp__synapse__synapse_scene_info,mcp__synapse__synapse_metrics,mcp__synapse__synapse_live_metrics). No doc reads beyond the §7 sequence you already carry in this charter. Numbers go to the file; your final text is the path + verdict only — never paste the histograms back.

Deliverable: write harness/notes/latency_measure_<YYYY-MM-DD>.md — the raw ping series (min/median/max), read-op timings, the four histogram snapshots verbatim, the skipped-step ledger, and a 3-line verdict on the one open dispute (does the panel path pay the 1–70 ms class or a dispatch floor — decided by dispatch_wait count vs main_thread_direct count). Return the file path plus the verdict as your final text. If a live second measurement file for today already exists, append a clearly-separated second run — never overwrite (Article V).

Changes

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.

  1. 6d ago First seen · 35 lines · 64 tokens per session scan A 42ad11ceea56

Subscribe to this mod's changes

latency-measurer is an agent published in the GitHub repository JosephOIbrahim/Synapse (10 stars, last pushed today), licensed MIT. It adds 64 tokens to every session and 637 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.