Two long games reached effectively the same position but finished with different scores: 1,811,320 vs 1,810,888, a gap of 432. The boards looked the same, so the scores "should" have matched. They don't, and there is no bug. Here is the full explanation, with the exact numbers.
There is a moment where both games show an identical set of 16 tiles (total
The usual MCP failure mode is a tool that returns a large value: the agent
either floods its context window or has to re-run the command with | head
to claw it back. Both options cost turns and lose information.
nu-mcp sidesteps this by storing every evaluation's full result in a ring buffer the agent can refer to by index on subsequent calls.
| ( | |
| interleave | |
| { generate {|_ = 0| | |
| let ev = (try { input listen --types [key] --timeout 200ms } catch { null }) | |
| {out: (if $ev != null { {kind: key, event: $ev} }), next: 0} | |
| } | |
| } | |
| { generate {|_ = 0| sleep 1sec; {out: {kind: tick, event: (date now)}, next: 0} } } | |
| ) | |
| | where $it != null |
| with-dotenv { | |
| echo '{"role":"user","content":"Run this in the nu tool exactly: use greeter; greeter hi"}' | |
| | yoke --provider openrouter --model openai/gpt-oss-20b:free --tools nu -I ~/yoke-test/modules | |
| } | |
| {"type":"tools","tools":[{"name":"nu","description":"Execute a Nushell script. Output is auto-converted to nuon -- do NOT add '| to nuon' or '| to json'. Pass structured data via 'input' (JSON) to avoid quoting issues -- it becomes $in in the pipeline.\n\nUse help liberally to learn how nushell works. Do NOT guess command names or flags -- discover them with help. If you encounter an error, you MUST use help related to your task before trying again.\n\nExamples:\n {command: \"help where\"}\n {command: \"help --find convert\"}\n {command: \"$in | where price > 20\", input: [{name: \"Widget A\", price: 25.50}, {name: \"Gadget\", price: 15}]}\n {command: \"seq 1 10 | each { |n| $n * $n }\"}"}]} | |
| {"type":"agent_start"} | |
| {"type":"turn_start"} | |
| {"role":"user","content":[{"type":"text","text":"Run this in |
In the for_collect case, my understanding is the main goal of FilePath is to prevent clobbering files with save. If we've collected, it's now safe to overwrite our original file, so FilePath is cleared. The code has always been doing this, but the previous version tied that decision to content_type — it only cleared FilePath when content_type was also None. That coupling didn't make sense, so when I added the custom metadata field, I moved the collect-time cleanup into its own for_collect method on PipelineMetadata and made it always clear FilePath regardless of other fields.
Stellar is a local-first CSS generator with an interactive web editor.
stellar serve serves your generated CSS and exposes a /live-refresh
SSE endpoint that pushes stylesheet updates via Datastar.
stellar serve # http://localhost:7331
Exploring schema validation for nushell/http-nu
We want to validate records against a schema (like JSON Schema). Nushell already has type syntax for command signatures:
def foo []: record<
name: string,
age: int,
email?: string,
tags: list<string>,| use xs.nu * | |
| use http-nu/router * | |
| use http-nu/datastar * | |
| use http-nu/html * | |
| def escape []: string -> string { | |
| $in | str replace -a '&' '&' | str replace -a '<' '<' | str replace -a '>' '>' | |
| } |