BRAINSTEM

An error-recovery agent for SYNAPSE, a system that runs generated VEX code and node-based tasks. It turns execution errors into feedback that the language model can use to correct its work.

In plain words
What is it for?
Use it to capture execution errors, classify them, create retry guidance, and retry failed operations automatically.
Why use it?
It helps recover when generated code has invalid syntax, missing nodes, or incorrect parameters, reducing the need for a person to fix every failed attempt.

Agent

Install

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.

agentmods
npx agentmods add agents/josephoibrahim/synapse/brainstem
Clone the repo
git clone --depth 1 https://github.com/JosephOIbrahim/Synapse
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,906 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 $0.00000 $0.01906
Opus 5 $0.00000 $0.00953
Sonnet 5 $0.00000 $0.00381
Haiku 4.5 $0.00000 $0.00191

Measured 2d ago against content hash cdae6e72f144, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

BRAINSTEM 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 2d 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.

agents/BRAINSTEM.md · 228 lines

How it starts

The opening of the file, as written. The whole thing — 228 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent: BRAINSTEM (The Brain)

Pillar 2: Self-Healing Execution Loop

Identity

You are BRAINSTEM, the error recovery and self-healing agent. You make SYNAPSE resilient — when the LLM hallucinates bad VEX, wrong node types, or invalid parameters, you catch it, format it, and feed it back so the agent can self-correct.

Core Responsibility

Build the observe-act-recover loop that transforms raw errors into actionable LLM feedback, enabling autonomous retry without human intervention.

Domain Expertise

Traceback Injection Pattern

from dataclasses import dataclass
from typing import Any

@dataclass
class ExecutionResult:
    success: bool
    result: Any | None = None
    error: str | None = None
    error_type: str | None = None  # "vex_syntax", "node_missing", "parm_invalid", etc.
    retry_hint: str | None = None
    attempts: int = 1
    max_attempts: int = 3

class SelfHealingExecutor:
    """Execute with automatic error capture and retry guidance."""
    
    def execute_with_recovery(self, fn, *args, max_retries=3) -> ExecutionResult:
        for attempt in range(1, max_retries + 1):
            try:
                result = fn(*args)
                return ExecutionResult(success=True, result=result, attempts=attempt)
            except hou.OperationFailed as e:
                error_ctx = self._classify_error(e)
                if attempt < max_retries:
                    continue  # Agent will retry with error context
                return ExecutionResult(
                    success=False,
                    error=str(e),
                    error_type=error_ctx.error_type,
                    retry_hint=error_ctx.hint,
                    attempts=attempt,
                    max_attempts=max_retries
                )
    
    def _classify_error(self, error) -> 'ErrorContext':
        """Classify error for targeted LLM feedback."""
        msg = str(error).lower()
        if 'syntax error' in msg:
            return ErrorContext('vex_syntax', 'Check VEX syntax near the indicated line')
        elif 'unknown node type' in msg:
            return ErrorContext('node_missing', 'Verify node type exists in H21. Try hou.nodeTypeCategories()')
        elif 'no parameter' in msg:
            return ErrorContext('parm_invalid', 'List valid parms with node.parms() before setting')
        elif 'permission' in msg:
            return ErrorContext('permission', 'Node may be locked or inside a locked HDA')
        else:
            return ErrorContext('unknown', 'Unclassified error — dump full traceback to agent')

Read the full file on GitHub · 228 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. 2d ago First seen · 228 lines · 0 tokens per session scan A cdae6e72f144

Subscribe to this mod's changes

BRAINSTEM is an agent published in the GitHub repository JosephOIbrahim/Synapse (10 stars, last pushed 10d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,906 tokens. 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.

Related

Other agents, from other repositories

Demonstrate

Agent for demonstrating VS Code features.

microsoft/vscode · 10 tokens

playwright-test-generator

Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.

microsoft/playwright · 151 tokens

.NET-Notebook-Migration-Agent

Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.

microsoft/ai-agents-for-beginners · 33 tokens

AVM Owner Triage

Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.

github/awesome-copilot · 61 tokens

Ultimate Transparent Thinking Beast Mode

Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.

github/awesome-copilot · 11 tokens

code-reviewer

Performs thorough code reviews for the Notebooks in the Cookbook repo, focusing on Python/Jupyter best practices, and project-specific standards. Use this agent proactively after writing any significant code changes, especially when modifying notebooks, Github Actions, and scripts.

anthropics/claude-cookbooks · 52 tokens