9 posts tagged with "Doctrine"

An executor's reasoning, false starts, and quoted templates used to ride along with everything else in its reply — all the way back into the coordinator's own context, and from there into the next task's prompt. Only the structured footer does that now; the rest is archived to disk and discarded from context.

Mythos-core rule 6 has always told a model to stop after two failed fix attempts and hand the problem up. That only works if the model is still counting straight — and a resumed or re-invoked task starts a fresh context with no memory of the attempts a prior run already burned. The orchestrator now keeps that count itself.

Anchor's hard rule used to read: run /commit-prep before any commit. /review had been ignoring it for months.

Anchor already stopped too-hard work with a first-line SUGGEST-ESCALATE. Premium sessions still burned credits on too-easy work with only a soft “say so and ask.”

Anchor’s fit check used to answer one question well: is this model strong enough? Mythos-core rule 11 and model-fitness.md sent weak-column and orchestration-class work up with SUGGEST-ESCALATE. That stopped silent overreach — but it did not catch the other failure mode: a model that is strong enough yet the wrong kind of product for the job.

A task that runs out of context used to fail the worst way available: a truncated answer, a confident "the rest is straightforward", or a silent stop that looks like success. Executors now emit a structured handoff as they approach their declared budget, and the orchestrator respawns a fresh context seeded with it.

Until now, only a /review survey could land a branch on dev. That rule is amended: a /work session may merge its own finished work onto integration — when the operator answers its end-of-run question in-session, and the branch passes a mechanical scoped-merge gate.

The best feature in your toolchain is worthless if nobody knows the command for it. A new doctrine rule closes that gap: before acting on a request, an Anchor agent checks whether a skill or slash-command already available in the session would do the job faster or more correctly — and if one clearly fits, it says so in a single line and offers to use it, then gets on with the work.

Two scattered disciplines — "will this even fit?" and "did I check the basics before diving in?" — are now one mechanical gate: every task spec declares a token budget up front, and mythos-core rule 13 makes every executor print a fixed pass/fail checklist before doing any work.