aide-arch-model-runtime SKILL.md skillA
aide-arch-model-runtime from anonymousnomad/aide-sovereign-workbench is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add anonymousnomad/aide-sovereign-workbench/aide-arch-model-runtime@git:20260823.df7cbdePins git:20260823.df7cbde in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- anonymousnomad/aide-sovereign-workbench
- Path
skills/packs/aide-arch-model-runtime/SKILL.md- Versions recorded
- 1
- Latest version
- git:20260823.df7cbde, committed Aug 23, 2026
- Size
- 12,722 bytes
- Repository stars
- 53
- License
- Apache-2.0
Phase 6 SOP for the AIDE offline-first IDE rebuild — the local model runtime + chat: port the VERIFIED model-manager (python discovery chain, llama_cpp.server spawn with --logits_all false, identity check, warmup gate, RAM doctrine, port doctrine 8081→8087), typed chat contract with SSE token streaming, model status via WS, offline-first (local llama.cpp only; online providers strictly opt-in). Use whenever wiring model start/stop/status/chat, debugging "model won't start", "first chat fails…
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# AIDE Architecture — Phase 6: Model Runtime + Chat ## Doctrine - **Local-first, always.** The 3 bundled GGUF models (smollm2, qwen-0.5b, qwen-1.5b) serve on localhost via llama_cpp.server. Chat never leaves the machine. Online providers (if ever added) are a SEPARATE opt-in adapter behind the same chat contract, defaulting to off. - **Verified facts are law.** The Phase 6 code that survived live verification (python discovery chain, logits_all false, identity check, warmup gate, RAM doctrine, port doctrine) is PORTED, not redesigned. New work extends the contract layer and streaming; it does not touch verified behavior. - Verify before claiming: every model start → status → chat round-trip verified live end-to-end (as done 2026-08-16) before the phase is done. ## What Phase 6 turns the verified model-manager.mjs into a first-class contract service: - **Model service** (`node/src/services/models.ts`): registry of models (id, path, port, status), start/stop/status, warmup gate, RAM guard, port allocation, python discovery — all verified logic ported from `daemon/model-manager.mjs`. …
Read the whole file at its exact version.
How to install
mdr add anonymousnomad/aide-sovereign-workbench/aide-arch-model-runtime@sha256:e8eace616ef788ddThe line at the top of this page pins the label git:20260823.df7cbde, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_n2j7rtdakdtqf3ft)
4 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 1 version of this file, committed on Aug 23, 2026.
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (12722 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
anonymousnomad/aide-sovereign-workbench · 53 stars · license Apache-2.0 · pushed 2026-10-26 · branch covert-production
API
GET https://markdownregistry.com/api/v1/artifacts/art_n2j7rtdakdtqf3ft GET https://markdownregistry.com/api/v1/resolve?ref=anonymousnomad/aide-sovereign-workbench/aide-arch-model-runtime GET https://markdownregistry.com/api/v1/blob/e8eace616ef788dd6a48988fb5042e0e4ab84a7bfbb2b245a7fe34aaa649bb69
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.