An independent, evidence-checked field guide to Jev
31 build blueprints drawn from the public record, 9 claims sorted by what has been demonstrated and what is still a vendor claim, and an MCP endpoint your coding agent can read directly.
Of 9 tracked claims, 3 are demonstrated by located work and 2 remain vendor claims. Status is recorded per claim and never upgraded by repetition.
Claims ledger →Nine lenses · what a workload needs before Jev fits
Open the opportunity map →01 Agent interface · MCP
The same normalized claims, projects, patterns, opportunities, evidence status, and sources that power this website, over one public read-only endpoint.
See the full agent interface →Without the atlas
Your agent answers from whatever it absorbed before Jev existed, or reads vendor pages and cannot tell a located demonstration from launch copy.
With the atlas
assess_jev_fitIs this workload actually a bounded semantic decision, and where does it stop?get_build_blueprintOpen an authored opportunity with its architecture, MVP, experiment, and unknowns.trace_jev_claimCheck a claim's evidence status, counterarguments, open questions, and sources.{
"mcpServers": {
"jev-atlas": {
"url": "https://YOUR-DOMAIN/mcp"
}
}
}https://YOUR-DOMAIN/mcpClient configuration formats vary. Use this endpoint anywhere your agent accepts a remote Streamable HTTP MCP server.
02 Field synthesis
What Jev is, what TypeSafe claims, what developers have actually built, and where the architecture gets interesting.
Read the full report →MEDIUM confidence / indie fit 10 of 10
Find a thread
Claims, projects, product ideas, source posts, and long-form synthesis are indexed together.
The research run
Three capped collection passes gathered the launch-period conversation, then a local pipeline classified and synthesized the retained material.
Costs are estimates based on provider pricing at collection time. They cover X API retrieval only, not analysis or development time.
03 Located work
Public repositories, demonstrations, and integrations—kept separate from proposals and launch copy.
Decision models can act as a semantic layer inside an existing event-driven rules engine.
Open ↗02The probability-bearing result can be preserved through agent infrastructure instead of flattened to prose.
Open ↗03Cheap typed judgments invite ensemble and quorum experiments, though correlated errors still need measurement.
Open ↗04Keeping selection probabilistic and execution deterministic makes the boundary easy to inspect.
Open ↗05A decision model can be invisible infrastructure beneath a workflow product.
Open ↗View the project index →About & methodology
The pipeline preserves sources, separates evidence levels, and keeps unanswered questions visible next to conclusions.
The complete view: evidence, developer activity, architectures, limitations, and opportunities.
Facts, vendor claims, independent observations, opinions, speculation, and unknowns.
Important claims with status, evidence, counterarguments, sources, and open questions.
Located projects and demonstrations, kept separate from proposals and speculation.
This is a launch-period, English-only, query-conditioned sample. Heuristic scores organize attention; they are not objective quality labels. The local dataset contains 0 processed records.