coreledger.ca

What It Is

Coreledger Technologies’ flagship product, Contextus, is positioned as a “control plane for AI Agent actions.” It is not a chatbot or foundation model; instead, it focuses on governance around developer agents’ context, tool calls, approvals, evaluations, and audits. It is aimed at teams that have already allowed agents to interact with real code, systems, or customer data, addressing questions such as: what did the agent see, what did it do, who approved it, and can this be proven afterward?

Core Capabilities

The product is organized around four pillars: Compile, Govern, Approve, and Prove. Compile gathers relevant files, documentation, policies, and task history before an action is taken, and references SitEmb, a context-aware ranking engine designed to reduce noisy context. Govern classifies high-risk tool calls such as writes, deletions, sending messages, deployments, network calls, and credential access according to policy. Approve pauses execution and requires human decision-making when risk increases. Prove preserves the context, policy, approver, reasoning, eval results, and audit records. Contextus also supports MCP, IDE/editor-native workflows, CI, and custom engineering environments, with API access available starting from the Builder plan.

Pricing and Trial

Pricing is publicly listed. The Free plan is $0/month and lets users try the core governance loop, with limited IDE/MCP tools, templates, 7-day data retention, and community support. Builder is $19/month and is aimed at individual developers, including personal audit history, Policy Editor, basic proof exports, API access, and eval reports. Team is $99/month and includes 5 seats, shared approval flows, audit exports, shared dashboards, and 90-day retention, with additional seats at $20 each. Enterprise is custom-priced and offers advanced approvals, custom retention, SSO/SAML/SCIM, security reviews, and a dedicated success manager.

Pros and Cons

The main advantage is its clear product focus: it targets the governance problems that emerge after AI agents are deployed in real workflows, combining context selection, policies, human approval, evals, and audit evidence into a single process. It also explicitly states that customer data is not used to train models. The limitations are also clear: “Request access” appears in multiple places, suggesting the product is still relatively early-stage; self-hosting and private deployment are only listed as planned items; and the specific IDEs, MCP hosts, LLM providers, and enterprise SLA details require direct confirmation. There is no visible disclosure of a Chinese interface, Chinese documentation, or China-region payment options.

Who It’s For and Access from China

Contextus is best suited to AI engineering teams, platform teams, and developer tooling teams, especially in scenarios where agents can already call tools, modify code, or affect production systems. Users focused only on content generation or lightweight automation may not need it. The site does not state its accessibility from China. Payments are processed by PCI-compliant services such as Stripe, so companies in China should independently verify network connectivity, foreign-currency card payments, compliance requirements, and cross-border data transfer considerations. If a localized alternative is required, teams should prioritize agent gateway or audit governance solutions that support private deployment, Chinese cloud providers, and Chinese-language support.