Castra
Strip orchestration rights from your LLMs
Details
- External ID
- 47601608
- Source
- HN
- Company
- —
- Product
- Xastra
- Website domain
- github.com
- Launched
- April 1, 2026
- Cohort
- —
- Upvotes
- 8
- Upvotes percentile
- 0.4813624678663239
- Tags
- —
- Fetched at
- Sept. 7, 2026, 9:26 p.m.
- Updated at
- Sept. 7, 2026, 9:26 p.m.
Description
I got tired of AI agents forgetting what they were doing the moment their context window filled. The current industry solution is to write massively bloated agent harnesses full of defensive spaghetti just to stop models from drifting.The problem is treating chat history as project state. A conversation is not a ledger.Castra is a compiled Go binary that strips orchestration rights from the LLM. State lives in an encrypted, local SQLite database (castra.db). The LLM is just a stateless executor — it reads the DB, executes a highly constrained task, and the result is written back subject to rigid state-machine rules.What it actually does: - 7-Role RBAC: Hard jurisdictional boundaries (Architect plans, Engineer builds, QA tests). - Dual-Gate Approval: A task cannot reach done without explicit, sequential approval from both a QA agent and a Security agent. No self-approving code. - Cryptographic Audit Chain: Every action is logged into a SHA-256 hash-linked, Ed25519-signed ledger. - Multi-Vendor: Works with Claude, Copilot, Gemini, etc. via a standard AGENTS.md protocol. Anything that supports AGENTS.md and can run terminal commands.Proof of Work: I built this by hand up to v1.3.0. Then, I turned Castra on itself. The agents governed by this exact CLI took over and built the architecture up to v3.1.2—including the cryptographic log chain itself. The proof is in castra-log.jsonl in the repo.If you are running multi-agent workflows and hitting the context amnesia wall, stop trying to prompt-engineer your way out of it. Fix the state machine.
Enrichment
- Theme
- AI agent frameworks and developer tools
- Vertical
- Horizontal
- Function
- Agent / copilot
- Audience
- Developer
- AI stance
- AI feature
- Project type
- Commercial product
- Normalized one-liner
- remove orchestration capabilities from llms
- Manually corrected
- False
Could you build this?
Yes It is a deterministic state machine and execution harness that strips multi-step routing authority away from the LLM, easily built with standard TypeScript/Python workflow orchestration libraries.
Discussion
9 comments analyzed.
Competitors mentioned: File-based memory (Markdown/Artifacts), Orchestrator agent with dynamic routing, Behavior trees
Concerns raised: Why is cryptography necessary for AI-generated proofs, Market relies on two main alternative approaches already, State machine trades dynamic flexibility for rigidity
Feature requests: Integration with FaultWall for data plane governance, Dual approval gates documentation/clarification
Competitors
Other products that read as similar to this one — 89 launches clear the similarity bar, closest 8 shown.
Attention rank: #49 of 90 (itself plus its competitors, highest first — normalized so YC and Product Hunt are compared fairly).
Launched 146 days after the earliest competitor.
- Turn wire protocols into structured statements LLMs can understand · hn · 2026-09-10 · 6 upvotes · similarity 0.48
- Castari – The runtime for agents that need a computer · yc · 2026-02-17 · 5 upvotes · similarity 0.42
- VoicEra · ph · 2026-09-18 · 1 upvotes · similarity 0.40
- Limits · hn · 2026-02-25 · 9 upvotes · similarity 0.40
- I built a firewall for agents because prompt engineering isn't security · hn · 2026-01-19 · 7 upvotes · similarity 0.40
- AgentAct AI · ph · 2026-09-30 · 1 upvotes · similarity 0.38
- Open-Source Article 12 Logging Infrastructure for the EU AI Act · hn · 2026-03-03 · 42 upvotes · similarity 0.38
- Plano · hn · 2026-01-06 · 8 upvotes · similarity 0.38
Other launches for this product
Same idea, different domain
Nobody's really built a agent / copilot tool for Agriculture yet.