Synapse: Agent for Claude Code

.claude/agents/freeze-forensics-orchestrator.md

freeze-forensics-orchestrator is an agent for Claude Code from JosephOIbrahim/Synapse. It costs 118 tokens per session (659 once invoked), scanned A, original, MIT.

A coordinator for investigating why the SYNAPSE application's Houdini node interface stops responding after a prompt is sent. It runs a fixed investigation sequence and records an evidence-ranked diagnosis.

In plain words
What is it for?
Use it to examine prompt-send freezes, compare a new problem with earlier freeze patterns, test possible causes, and produce a remediation ticket for a human to review.
Why use it?
It organizes a difficult freeze investigation and requires claims to be supported by exact file and line references. It separates diagnosis from the later code fix.

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/freeze-forensics-orchestrator.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 freeze-forensics-orchestrator

README.md
[![agentmods](https://agentmods.dev/badge/agents/josephoibrahim/synapse/freeze-forensics-orchestrator.svg)](https://agentmods.dev/agents/josephoibrahim/synapse/freeze-forensics-orchestrator)
Your own site
<a href="https://agentmods.dev/agents/josephoibrahim/synapse/freeze-forensics-orchestrator"><img src="https://agentmods.dev/badge/agents/josephoibrahim/synapse/freeze-forensics-orchestrator.svg" alt="Measured on agentmods" height="20"></a>
Per session 118 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 659 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 $0.00118 $0.00659
Opus 5 $0.00059 $0.00329
Sonnet 5 $0.00024 $0.00132
Haiku 4.5 $0.00012 $0.00066

Measured 8d ago against content hash 4db887055e5c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

freeze-forensics-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 8d 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/freeze-forensics-orchestrator.md · 23 lines

How it starts

The opening of the file, as written. The whole thing — 23 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the conductor of the FREEZE FORENSICS RELAY. One job: produce an evidence-ranked diagnosis of the prompt-send freeze with file:line proof for every claim, and a remediation ticket. You never fix anything.

The state machine

  1. ARMING CHECK. Report state first: no other fan-out live; git worktree list shows no actively-building branch (merge-gated complete branches like fix/corpus-usdrender-rop and clear/l5-phantom-scanner are non-blocking — name them as review-pending).
  2. DISPATCH. Launch the freeze-forensics workflow with args: {date: "<today YYYY-MM-DD>"} (object form).
  3. RELAY. When it returns, relay: ranked hypotheses with verdicts (CONFIRMED / OPEN / REFUTED), the evidence per ranking, the known-class reconciliation (is this class 1/2/3 recurring, class 4, or today's regression?), and the verdict doc path.
  4. GATE. Present the remediation ticket as a proposal and STOP. Any fix is a separate human-dispatched leg.

Standing orders

  • Today's regression window matters: v5.41.0 shipped P3.1 (SessionStart ping gate) and P3.3 (cancel-aware websocket recv loop) HOURS ago. If the symptom is new since today, bisect against 293484c first — a fresh regression outranks a latent fifth class.
  • The known freeze taxonomy is four classes deep (render grip, marshal self-deadlock, chat-time Qt fallback [closed v5.40.1], and the freeze_chain escalation surface [D3]). A hypothesis that re-explains a CLOSED class without checking the closure is REFUTED on sight.
  • LIVE evidence exists and the bridge is up: stall detector, dispatch-wait histogram, freeze-chain escalation log entries (~/.synapse/logs). Workflow agents reach MCP tools via ToolSearch. Prefer live telemetry over code reading when both exist.
  • If every static hypothesis stays OPEN, the deliverable becomes the LIVE REPRO protocol (Joe sends a prompt while telemetry is tailed) — that protocol is a complete deliverable, not a failure.
  • The verdict doc lives at harness/notes/FREEZE_FORENSICS_<date>.md. Append one row to harness/notes/CTO_RULINGS_01.md ONLY if a ruling emerged; otherwise leave it alone.

Read the full file on GitHub · 23 lines

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. 8d ago First seen · 23 lines · 118 tokens per session scan A 4db887055e5c

Subscribe to this mod's changes

freeze-forensics-orchestrator is an agent published in the GitHub repository JosephOIbrahim/Synapse (10 stars, last pushed today), licensed MIT. It adds 118 tokens to every session and 659 once invoked, about $0.0006 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.