Valstorm Agent Runtime (vsagent)
Enterprise-grade autonomous AI execution runtime, ReAct loop, and parallel subagent swarm engine for developer workstations, CI pipelines, and remote team automation.
Run vsagent locally from your terminal, launch it as a background workstation daemon controllable from your phone, or execute generated code inside isolated Docker and E2B sandboxes. Verified across 6 scientific ablation axes with a 64.8% token reduction via persona-scoped tool registries and up to 3.00x parallel subagent swarm speedup.
1. Interactive REPL, Single-Shot Execution & Parallel Subagent Swarms
# 1. Install vsagent globally via uv
uv tool install --from git+https://github.com/Valstorm-Corp/valstorm-agent.git valstorm-agent --force
# 2. Launch interactive multi-turn REPL with a specialized profile
vsagent chat --profile developer --model gemini-flash-latest
# 3. Execute a headless single-prompt task in CI or local terminal
vsagent chat -p "Run pytest and fix any failing assertions" --profile backend-tester
# 4. Run the 6-axis scientific benchmark & ablation harness
uv run python -m benchmarks.runner --ablation all --mode mock
2. How valstorm-agent Works Under the Hood
📱 Mobile App / Web Browser / CLI Terminal
│
▼ (Encrypted WebSocket / REST API)
☁️ Valstorm Enterprise Cloud Gateway (Optional)
• Multi-Tenant State & Audit Logging
• Managed Multi-Model Gateway (Gemini, Claude, OpenAI, DeepSeek)
• Knowledge Vaults (VFS) & 170+ Procedural Skills
│
┌─────────────┴─────────────┐
▼ ▼
💻 Local Workstation 🛡️ Ephemeral Cloud Sandboxes
• vsagent Daemon (8650) • Isolated Docker / E2B
• Scoped Tool Registries • Parallel Asyncio Subagent Swarms
• Git repos & Local Shell • Human-in-the-Loop Approval Gates 3. Core Capabilities & Technical Specifications
Persona-Scoped Tool Registries (-64.8% Tokens)
Instead of dumping 30+ tool schemas into every prompt, vsagent scopes tool access per profile (developer, researcher, archivist, backend-tester), cutting token usage by 64.8% and inference costs by 2.8x.
Non-Blocking Parallel Subagent Swarms (background=True)
Spawns isolated child subagents via asyncio.create_task in under 70ms and awaits completion with asyncio.gather(), compressing wall-clock execution time to the slowest single worker.
170+ On-Demand Procedural Skills (SKILL.md)
Loads battle-tested standard operating procedures (TDD, bash-scripting, systematic debugging) on demand to eliminate operational regressions like missing chmod +x permissions.
Human-in-the-Loop & Ephemeral Sandboxes
High-risk operations automatically pause via confirmation_required, while untrusted code runs inside isolated Docker containers or E2B sandboxes.
4. Open-Source & Commercial Licensing
PolyForm Noncommercial License 1.0.0
Free to clone, inspect, modify, and run for personal projects, educational study, and noncommercial research under the PolyForm Noncommercial License 1.0.0.
Valstorm Commercial License
Commercial use in business operations is included automatically with any active Valstorm BOS workspace ($25/mo in usage credits with $0/seat tax) or available via enterprise BYOC licensing.
5. Frequently Asked Questions
Can I run valstorm-agent locally with my own LLM API keys?
Yes. Under the PolyForm Noncommercial License 1.0.0, you can install vsagent via uv and configure your own Gemini, Anthropic, OpenAI, or DeepSeek API keys for personal, academic, and noncommercial research use.
How do I use valstorm-agent commercially across my company?
Commercial use is included automatically with any active Valstorm BOS workspace ($25/month in usage credits with $0/seat fees), which also gives your team managed multi-model billing, cloud sandboxes, and mobile-to-workstation remote control.
Explore Other Valstorm Open-Source Packages
View All 7 Tools →Command-line interface for Valstorm developers to authenticate via OAuth2 PKCE, pull and push object schemas as code, and manage local workspaces.
Stateless Streamable HTTP, SSE, and stdio Model Context Protocol (MCP) server exposing 16 granularly scoped Valstorm VFS, SQL, and CRM tools.
High-performance web scraping, domain crawling, RSS monitoring, and visual diff microservice with Python 3.14 native Zstandard compression.
Dedicated asynchronous microservice for multi-preset FFmpeg video transcoding, Deepgram/Whisper speech-to-text captions, and AI chapter generation.
Official typed Python 3.13+ SDK with 1:1 PlatformContext parity across records, SQL/GraphQL queries, VFS Knowledge Vaults, schemas, and sandboxes.
Pure Python standard-library secret redaction engine that strips API keys, JWTs, database passwords, and environment secrets before logs or LLM prompts egress.