Auto Model Router
v1.0.0Auto Model Router - Automatically select the best model based on task complexity for OpenClaw agents
⭐ 0· 68·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
Capability signals
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
OpenClaw
Benign
medium confidencePurpose & Capability
Name/description (route tasks to models) matches files and SKILL.md: task categories, model plans, config files and mapping rules are all about model selection and sub-agent dispatch. There are no unrelated env vars, binaries, or config paths requested.
Instruction Scope
SKILL.md contains detailed classification logic, config locations (~/.openclaw/skills/auto-model-router/), and steps to save config and 'auto reload OpenClaw'. It also describes generating sub-agents and querying model capability via 'local database' or 'API identification' but does not include concrete commands or code. The instructions are scoped to routing/config; they do instruct writing/reading skill-specific config in the user's OpenClaw folder (expected) and leave implementation details vague.
Install Mechanism
No install spec and no code files to install — instruction-only skill, so nothing is downloaded or executed by a packaged installer. This is the lowest-risk install pattern.
Credentials
Skill declares no required environment variables or credentials. The SKILL.md references many third-party model identifiers (OpenAI, Anthropic, GLM, etc.); using those providers in practice requires API keys/credentials elsewhere on the agent/platform. The lack of explicit credential requests is coherent for a reusable skill, but users should understand provider/API credentials must be present in their environment for routing to be effective.
Persistence & Privilege
always is false and the skill is user-invocable. The skill expects to persist its own config under ~/.openclaw/skills/auto-model-router/, which is normal for a skill. It does not request system-wide changes or other skills' credentials.
Assessment
This skill appears to be a documentation/config helper that tells OpenClaw how to map tasks to models. It does not bundle code or request credentials, so installing it won't itself send secrets anywhere. Two practical things to consider before installing: (1) To actually call the listed models (OpenAI, Anthropic, GLM, etc.) your agent environment must already have the appropriate API keys configured — the skill does not supply or request them. (2) The SKILL.md is a high-level specification (writes config under ~/.openclaw/skills/auto-model-router/ and mentions auto-reload and sub-agent creation) but provides no concrete code; runtime behavior depends on your OpenClaw agent implementing those actions. If you want fully automated routing behavior, confirm your agent/platform implements the config write, reload, and sub-agent creation steps, and review any runtime code that will act on these config files before enabling autonomous invocation.Like a lobster shell, security has layers — review code before you run it.
latestvk9792fag9yrdawh92e8388s2p984dpa7
License
MIT-0
Free to use, modify, and redistribute. No attribution required.