Overview
OneCLI addresses a critical and growing security gap: AI agents need credentials to perform real tasks, but giving them direct access introduces severe risks like prompt injection and credential leakage. By acting as a network-level credential gateway, it enforces policies without requiring agent trust. The solution is open-source, works with any agent framework, and supports human-in-the-loop approval. This is a well-timed, high-value offering for teams deploying AI agents in production, especially in security-conscious environments. The potential to become a standard security layer for agent ecosystems is significant.
Score Breakdown
| Advisor 8.0 | Devil -6.0 | Historian 3.0 | Budget Steward 0.0 | Founder 7.0 | ⭐ Total Score 4.2/10
What They Do
hey HN, Jonathan and Guy here, creators of OneCLI ( https://onecli.sh/ ). OneCLI is an open source vault for AI Agents. Traditional vaults are used to store your secrets and, on demand, provide them to you all in a secure way, trusting the person to keep them safe. We figured that in the agent's world, this is not the case, as you don't know what happens with the secret after it's delivered to the agent, or where it was saved. Or maybe someone even manipulated them to hand them over... From that understanding, we decided to build a network gateway that sits between your AI agents and the services they call. OneCLI matches the request by host/path, verifies the agent should have access, swaps the placeholder for the real credential, and forwards the request. the secrets set inside the OneCLI vault, encrypted on rest, or could fetch in realtime from your bitwarden / 1password wallets. Demo - . We started working on this by accident, even though our careers were in the security space. We were working on a devtool called ChartDB, an open-source DB tool. When OpenClaw took off back in January, we started using it to orchestrate agents on top of ChartDB. We quickly understood there is a big issue around auth. Agents need credentials to do real work, but to give them those secrets would not be the best idea. they keep them in their memory and also write them down to local files and their sessions as plain text. And we knew that agents can easily be fooled into giving up those API keys/secrets. So we needed some way to control the agent and stop prompt injections from tricking it into using its services for an attacker's benefit. Not providing the keys to the agent + adding alerts or human-in-the-loop for sensitive operations, in case someone manipulates the agent, and access logs are being audited. We use it ourselves daily. My OpenClaw runs my day across Attio, Gmail, and my calendar, with human-in-the-loop approval on sensitive actions. The approval is enforced at the network layer, so it holds whether the agent goes through MCP, CLI, curl, or code it wrote on the fly. Guy uses it to review PRs but can't merge without his approval. Companies mostly use it for coding agents currently, which get creative about finding ways to elevate their permissions to reach the goal. With OneCLI, it doesn't matter what the agent finds: if the request isn't in policy, it will not go through. The proxy is written in Rust, the dashboard is Next.js, and secrets are AES-256-GCM encrypted at rest. Everything runs in a Docker container. It works with any agent framework (Claude Code, Codex, Cursor, OpenClaw, Hermes or anything that can set an HTTPS_PROXY). It won't stop an agent from misusing access it legitimately has, so scope policies tightly! happy to answer anything. We believe we can't trust the model to behave and have to set deterministic rules to stay in control. Comments welcome!
Why It Matters
OneCLI addresses a critical and growing security gap: AI agents need credentials to perform real tasks, but giving them direct access introduces severe risks like prompt injection and credential leakage. By acting as a network-level credential gateway, it enforces policies without requiring agent trust. The solution is open-source, works with any agent framework, and supports human-in-the-loop approval. This is a well-timed, high-value offering for teams deploying AI agents in production, especially in security-conscious environments. The potential to become a standard security layer for agent ecosystems is significant.
Business Model
To be determined based on project analysis.
CRP Decision Report
🧠 Advisor
OneCLI addresses a critical and growing security gap: AI agents need credentials to perform real tasks, but giving them direct access introduces severe risks like prompt injection and credential leakage. By acting as a network-level credential gateway, it enforces policies without requiring agent trust. The solution is open-source, works with any agent framework, and supports human-in-the-loop approval. This is a well-timed, high-value offering for teams deploying AI agents in production, especially in security-conscious environments. The potential to become a standard security layer for agent ecosystems is significant.
[score: 8]
[score: 8.0]
🧨 Devil
- Adoption friction: Setting up a proxy and defining policies adds complexity. Non-security teams may find it burdensome, limiting adoption.
- Bypass risk: If an agent can make direct network calls (e.g., via a library that ignores HTTPS_PROXY), the gateway is ineffective. The system relies on the agent cooperating.
- False sense of security: Misconfigured or overly permissive policies can create dangerous blind spots. Users might assume the proxy covers all threats.
- Performance overhead: All requests get proxied, introducing latency. For high-frequency agent tasks, this could be a dealbreaker.
- Competition: Integrated solutions (e.g., within OpenClaw, Cursor, or future operating system-level controls) could render OneCLI redundant.
[score: -6]
[score: -6.0]
📚 Historian
Historically, credential management tools like HashiCorp Vault, AWS Secrets Manager, and proxy-based security gateways have seen steady adoption but often struggle with complexity and integration friction. The pattern of "gateway before the agent" is novel, but similar attempts to secure API access at the network layer (e.g., API gateways) have succeeded in enterprise contexts. The current AI agent boom mirrors earlier shifts (e.g., cloud, mobile) where security tools emerged later to capture the market. OneCLI's timing is favorable, but the risk of being too early or too niche is real. Overall, the pattern is moderately positive, but execution and community adoption will determine success.
[score: 3]
[score: 3.0]
🧮 Budget Steward
This is an open-source project with a working Rust proxy and Next.js dashboard. To commercialize (first paying customer), we need to offer a hosted version or enterprise support. Estimated costs: $100/month for a small cloud VM (3 months = $300), domain and SSL ($50), third-party integrations (Bitwarden/1Password are free), legal template for terms ($500), and minimal customization (e.g., adding a billing layer, ~$500 if outsourced). Total ~$1,350. This is under $1,600, so +1 point. However, if we assume more robust hosting and compliance, it could exceed $1,600 but stay under $15,000 (0 points). Given the conservative estimate, I'll assign +1.
[score: +1]
🧱 Founder
OneCLI solves a real problem with a clean, pragmatic approach. The risks are manageable—focus on clear documentation, agent-agnostic design, and a hosted offering to reduce friction. The timeline to first paying customer is short (weeks, not months) because the core is already built. The recommended path: continue open-source development, offer a SaaS version with zero-setup, and target early adopters in security-conscious AI teams. Yes, we should build and scale this.
[score: 7]
[score: 7.0]
Why This Made Only 10
This project was selected because it scored highly across our evaluation framework. The CRP analysis confirmed strong potential, making it one of the most promising opportunities this week.