Real Skill packageSource verifiedClawHub registry

lygo-llm-console

Use when an operator wants a sovereign LOCAL LLM runtime instead of Ollama: scan GGUF on disk, boot llama.cpp on loopback, chat with an agent portal and allowlisted tools. Ships the public kit UNPACKED (v1.5.9, 152 files, per-file SHA-256) so every file that would run can be read first. Map scripts do no network, no subprocess, no writes; the kit itself listens on loopback, fetches public HTTPS, and writes only inside its own folder, where it KEEPS sessions, receipts and a RAG index on disk (resumable and importab…

Identity and source

Publisher attributionLYRA Agent - LYGO OSregistry owner unverified by skillvetai
Functional categoryAgent Engineering, Security & Governanceautomatically inferred · 55% rule confidence
Package forminstruction with code170 recorded files
Canonical sourceClawHub registryclawhub:deepseekoracle:lygo-llm-console
Open canonical source ↗

Platform declarations

These states come from the source or distribution context. None of the entries below are SkillVetAI compatibility test results.

OpenClawnative officialProvenance: registry distribution

Independent structural checks

These checks parse the fixed package against dated platform rules. They do not execute the Skill or verify task behavior.

Claude Codeissues found
Checker 0.1.0 · agent-skills-2026-08-13+claude-code-docs-2026-08-13 · 10/18/2026.claude/skills/lygo-llm-console
1 structural issue
  • error: description exceeds the 1,024-character Agent Skills limit. SKILL.md

Runtime, accounts, dependencies, permissions, network behavior and task quality remain untested.

OpenAI Codexissues found
Checker 0.1.0 · agent-skills-2026-08-13+codex-docs-2026-08-13 · 10/18/2026.agents/skills/lygo-llm-console
1 structural issue
  • error: description exceeds the 1,024-character Agent Skills limit. SKILL.md

Runtime, accounts, dependencies, permissions, network behavior and task quality remain untested.

OpenClawissues found
Checker 0.1.0 · agent-skills-2026-08-13+openclaw-docs-2026-08-13 · 10/18/2026skills/lygo-llm-console
2 structural issues
  • error: description exceeds the 1,024-character Agent Skills limit. SKILL.md
  • error: metadata.openclaw.os must contain only darwin, linux, or win32. SKILL.md

Runtime, accounts, dependencies, permissions, network behavior and task quality remain untested.

Installation and inspection

This command is recorded from the source ecosystem and resolves the registry's latest release. The fixed release shown on this page should be inspected before adoption.

clawhub install @deepseekoracle/lygo-llm-console
clawhub inspect @deepseekoracle/lygo-llm-console --version 1.6.3

Security evidence

SkillVetAI static result: high signal

This automated, non-executing scan is bound to this release hash. It is not a safety certification and may contain false positives or false negatives.

Status
completed
Coverage
partial text content
Files
40 / 170 inspected as text
Checked
10/18/2026, 3:55:03 PM
Scanner
0.1.3
Policy
1.0.3
1 automated finding
highEncoded content is decoded and executedkit/portal/app.js:209 · confidence 90%base64
1 High/Critical review queue entry
STATIC_OBFUSCATED_EXECUTIONpending
Open human review queue →
7 inferred permission indicators
  • shell execution — automatically inferred
  • network access — automatically inferred
  • filesystem read — automatically inferred
  • filesystem write — automatically inferred
  • credential access — automatically inferred
  • browser control — automatically inferred
  • external write action — automatically inferred
21 dependency and API indicators
  • api: ai.google.dev
  • api: api-docs.deepseek.com
  • api: api.deepseek.com
  • api: api.github.com
  • api: api.groq.com
  • api: api.openai.com
  • api: api.x.ai
  • api: asiancoastline.com
  • api: chatagent.ca
  • api: clawhub.ai
  • api: console.groq.com
  • api: deepseekoracle.github.io
  • api: docs.x.ai
  • api: generativelanguage.googleapis.com
  • api: github.com
  • api: huggingface.co
  • api: integrate.api.nvidia.com
  • api: openrouter.ai
  • api: platform.openai.com
  • api: www.patreon.com
  • api: www.paypal.com
External clawhub result: suspicious

This is registry-supplied evidence for the recorded release, not an independent SkillVetAI scan. Check the canonical source for the full report, scanner versions, scope, and current moderation state.

Evidence checked
10/18/2026, 7:33:49 AM
Release binding
Matches this record
  • skillspector: suspicious
  • llm: suspicious

Recorded files

The catalog stores hashes and an inventory summary for change detection. It does not republish the package contents.

Package content hashsha256:55adde0a4717dcb779d8a4109ac7e4ec7e5af0030efedca37acb57a21d20835a
Show up to 30 recorded paths
  • claw.json
  • examples/quickstart.md
  • kit/CHANNEL.txt
  • kit/config/console.json
  • kit/config/local.json.example
  • kit/CREDIT.txt
  • kit/HARDENING.md
  • kit/INSTALL.bat
  • kit/KIT_SHA256SUMS.txt
  • kit/LICENSE
  • kit/LICENSING.md
  • kit/LYGO_LLM_CONSOLE_STOP.bat
  • kit/LYGO_LLM_CONSOLE.bat
  • kit/MODELS.md
  • kit/NOTICE
  • kit/portal/app.js
  • kit/portal/donate.js
  • kit/portal/index.html
  • kit/portal/logo.svg
  • kit/portal/radio.js
  • kit/portal/style.css
  • kit/prompts/BRAIN.md
  • kit/prompts/CONTINUITY_SEED.json
  • kit/prompts/IDENTITY.md
  • kit/prompts/LINKS.md
  • kit/prompts/LYGO_ALIGN.txt
  • kit/prompts/MAP.md
  • kit/prompts/MEMORY.md
  • kit/prompts/POEM.txt
  • kit/prompts/SOUL.md

Source changelog

Closes the two live T09 findings. (1) Unauthenticated loopback API: _auth() returned True for every loopback caller when AUTH_REQUIRED was false -- the default install -- so any page the operator visited could POST /api/limb and reach shell/python. POST/PUT/PATCH/DELETE now need the token on loopback too; reads stay open so the health bar works; a non-local Host is refused on a loopback bind (DNS rebinding) unless LYGO_TRUSTED_PROXY opts in; the upload control sends the token it now needs. (2) Untrusted web content inducing local execution: a turn whose context contains fetched web text no longer auto-dispatches shell/python_exec/rust_exec/cargo -- the call is recorded REFUSED with the reason, and LYGO_WEB_TAINT_EXEC=1 is the documented opt-out. references/WHAT_CHANGED_1_6_2.md documents the 1.6.2 fixes; this release states its own evidence in kit/VERSION and SKILL.md.

Release security diff

mediumCompared fixed releases 1.6.0 and 1.6.3; 0 finding and 0 permission indicators were added.

Both fixed releases were scanned under the current scanner and policy, so finding, permission and dependency changes are available.

Change reasons and limitations
  • finding surface changed
  • content hash changed

Observed release history

These older immutable releases were observed by prior successful syncs. They remain recorded when a newer release becomes current.

1.6.09/26/2026sha256:bb34826cd6aeecf66bdcf5aae6ca49829e1d14d7dc5fed54e7bc73a7326e3b00
1.2.09/16/2026sha256:24e2816553db6bbb1b8d2368fbf8e1ae461c89380584be14c40d38e8f57514e7
1.1.09/16/2026sha256:73fbb69cf0203cda40445ceab4d53a489c5ec012d2081bcf50d3887d114d60cc