Loading Ashlr Brief
The briefing
AI-generated illustration
On October 25 Phantom added signed run receipts, and Ashlr's tools learned a shared handshake so each one's doctor can find and repair the others.
October 25 (UTC) brought about 185 merged pull requests across Ashlr's public repositories. Two ideas stand out.
First, Phantom run receipts. A receipt is a small, signed record of what an agent run did. The first pull request defines its shape and signs it with an Ed25519 key, and rejects anything unexpected. Secret names never enter a receipt; it holds only a count and a digest. Later the same day, Phantom added a local receipt store, wrote receipts for real dispatches, showed receipt and custody counts in phm status, added phm receipt timeline, and wired signed follow-up entries for CI, merge, survival and revert events. Receipts are opt-in for now.
Second, a shared handshake. Phantom, Locus, Lexicon, Ashlr MD, ashlr-plugin and Ashlr Stack each gained an --ashlr-handshake flag that reports the tool's version, health and capabilities in one agreed format. Ashlr Stack published the canonical contract and a checker the others vendor in tests. With it, stack doctor --ecosystem lists installed Ashlr tools, and when a tool advertises a doctor --fix capability, the others can suggest or preview its repairs. Locus, Lexicon and Ashlr MD all added doctor --fix, which previews safe repairs and applies them only with --apply.
Phantom Secrets continued hardening: commands now refuse a locked file vault with a clear message, state files are kept owner-only, and symlinked config files are refused. Phantom also added a local inference guide for Ollama, oMLX and mlx-lm, with a loopback-only setting for local models.
A small versioned handshake with a conformance test lets separate tools cooperate without tight coupling. If you ship several CLIs, one shared --json status contract is worth the effort.
Signed receipts move an agent's work from 'trust me' to a record you can check later. When evaluating agent tools, ask what record remains after a run and whether it can be verified.
If you use more than one Ashlr tool from source, run stack doctor --ecosystem to see how they report on each other.
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 AIThe briefing
Agentic AIAshlr release
Keep coding agents scoped to the right client, account, and tenant. Manage identity bindings across command-line tools and MCP connections.
Provision and connect development services through CLI and MCP tools, with Phantom Secrets handling supported credential paths.
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.
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.
