Graham grahama1970

External Review Prompt (Extractor — Canonical v1)

Goal: Deliver a blunt, evidence‑backed production‑readiness assessment of the Extractor project and a minimal patch set (unified diffs) with tests and doc updates. Keep changes surgical. Ship safety first.

Reviewer persona & tone

Principal SRE/DevEx + AppSec; fluent with Python/uv, Typer/FastAPI, Vite/React, ArangoDB, CI.
Be terse, specific, and fail‑closed. Unverified claims must be marked 🔴 (blocking) or 🟡 (needs proof). No hand‑waving.

Project context (declare at top of your report)

Project: Extractor — Self‑Correcting Agentic Document Processing System (multi‑stage pipeline + tabbed UX)

AGENTS.md

Repository Guidelines

Based on OpenAI Prompting Guide.

Agent Quickstart (Codex CLI)

Activation: Start with the prompt:
Activate the current dir as project using serena

Here is the final clear and concise estimate with math for how many full-time developers a 16 Nvidia H200 GPU cluster can support running Grok’s Kimi-k2, based purely on token throughput:

Given:

Average tokens per developer per day: ~5,800,000 tokens
Seconds per day: 86,400 seconds
Per-GPU token throughput (prefill + decode combined) from Grok Kimi-k2 benchmarks: ~4,000 tokens/sec
Number of GPUs in cluster: 16

Step 1: Calculate tokens per second per developer

name: code-reviewer description: Proactively triggered for comprehensive code reviews and security analysis. If they say 'review code', 'check security', 'code quality', or mention pull requests, use this agent. When you prompt this agent, provide the code location and review focus areas (security, performance, style). Remember, this agent has no context about your standards, so specify coding conventions and acceptable risk levels. tools: python type: reviewer capabilities:

code_review_bundle
git_context_integration
security_analysis
performance_review

Claude Dashboard

A real-time monitoring dashboard for Claude code instances managed by claude_dispatcher.py, built with React, Tailwind CSS, Shadcn/UI, and FastAPI. It provides live observability of tmux slots, Claude hook events, and system health, inspired by the disler/claude-code-hooks-multi-agent-observability repository. The dashboard runs locally, leveraging uv for dependency management and WebSockets for real-time updates.

Key Features

Real-Time Monitoring: Displays live tmux slot statuses, Claude hook events (e.g., PreToolUse, Notification), and system health (CPU, memory, Redis) via WebSockets.
Live Tmux Output Streaming: Streams tmux capture-pane output for each slot in real-time, viewable in a modal.
Advanced Event Filtering: Filters hook events by event type, session ID, source app, and content on the Event Timeline page.
Chat Transcript Modal: Views Claude conversation history for Notification events, with formatted prompt/response pairs.

	# ======================================
	# CORE
	# ======================================
	model = "gpt-5"
	model_reasoning_effort = "high"

	# Disable all sandboxing (no filesystem/network restrictions)
	sandbox_mode = "danger-full-access"

	# Never prompt for approvals (Codex will run commands directly)

	# codex_exec.py
	"""
	Async wrapper for running `codex exec ...` with robust timeout, streaming, and termination.

	Key features:
	- Overall and idle timeouts (wall and silence).
	- Graceful shutdown (SIGTERM) → hard kill (SIGKILL) with process-group awareness.
	- Stream readers that cannot deadlock; cancellation-safe finalization.
	- Rolling capture limits to avoid unbounded memory growth.
	- Optional binary or decoded text outputs.

	# --- Place below in .zshrc ---------

	# === Self-contained codexer with resume seed builder ===


	# --- codexer: simple conversation loader for the Codex CLI --------------------
	# Features:
	# --resume Append the last conversation (user + assistant) to the seed
	# --limit N Include only the last N lines of that conversation
	# --id SESSION_ID Resume a specific session id (instead of the most recent)

	#!/usr/bin/env python3

	"""
	LiteLLM Call - Easy async LLM batch runner with automatic image support

	WHAT IT DOES:
	- Run multiple LLM prompts in parallel for speed
	- Automatically detects and includes images from URLs or local files
	- Works with any LiteLLM-supported model (OpenAI, Anthropic, Ollama, etc.)
	- Handles all image processing automatically (compression, base64 encoding)

	#!/usr/bin/env python3
	"""
	Codebase Indexer for Semantic Code Search

	A tool for indexing code repositories into ArangoDB with semantic embeddings,
	enabling intelligent code search beyond simple text matching.

	Key Features:
	- Extracts functions/classes using tree-sitter AST parsing
	- Generates semantic embeddings using nomic-embed-code model (1024-dim)