Loading Ashlr Brief
The briefing
AI-generated illustration
Overnight into October 27, Phantom began recording each run's routing decision, Ashlr Stack added a merge-queue gate, and ashlr-plugin raised its MCP SDK floor.
In the early hours of October 27 (UTC), a handful of pull requests merged in Ashlr's public repositories. The most visible is in Phantom.
Phantom's hybrid local and frontier routing, added the day before, now keeps accounts. Each finished run records its routing decision in a local file with owner-only permissions. phm status --json shows the split between local and frontier work over the last 24 hours and the frontier tokens saved, and phm doctor gains a routing row. The description is explicit that this is accounting, not a signed receipt.
Ashlr Stack added a merge_group trigger and an aggregate CI gate, so its repository can use GitHub's merge queue with one required check. ashlr-plugin raised its minimum Model Context Protocol SDK version to 1.32.2. Ashlr Stack also merged several smaller maintenance changes.
Measure before you optimize. Recording each routing decision makes it possible to check whether local models are actually taking load, instead of assuming they are.
Savings claims are only as good as their accounting. A status view that shows the local and frontier split makes the cost story checkable.
If you try Phantom's hybrid routing, check phm status --json after a day of work to see the actual split.
Ashlr tools connected to this story. Review their source and setup requirements.
Work with eligible coding-agent accounts and local models in one workbench; review fleet proposals under signed authority.
Source material behind this briefing. Our practical implications are Ashlr’s interpretation.
This briefing was prepared with AI assistance from the linked sources. The analysis may include interpretation; consult the original sources for complete context. Cover artwork is illustrative.
Agentic AIThe briefing
AI-generated illustration
OpenAI launched GPT-6.1 Terra, a mid-tier model for computer use and visual reasoning tasks. It costs roughly half as much as GPT-5.6 Terra while running faster on agent workflows.
Agentic AIAshlr release
Agentic AIThe briefing
AI-generated illustration
OpenAI released GPT-6.1 Terra, a fast model for computer use and everyday work at $1 per million input tokens and $5 per million output tokens.
Provision and connect development services through CLI and MCP tools, with Phantom Secrets handling supported credential paths.
Token-efficient tools and workflows for Codex and Claude Code. Read, search, edit, and inspect repositories with compact outputs and visible context usage.
AI-generated illustration
October 26 brought releases of Locus 0.5.2, Ashlr Stack 0.2.2 and ashlr-plugin 1.37.1, plus opt-in routing in Phantom that tries local models first.