Hive Task Router
v1.0.0Intelligent task routing system that identifies task types (web/code/data/doc/chat) and routes to optimal models with appropriate execution mode (subagent/ma...
⭐ 0· 204·1 current·1 all-time
byqiongcao@qiongqiongcaocao
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description match the delivered artifacts: SKILL.md and router.sh implement keyword-based task classification, model selection, and spawning OpenClaw sessions. Required tooling (openclaw CLI, bash) and environment-variable model IDs are appropriate for a multi-provider router.
Instruction Scope
Instructions and router.sh focus on task classification, model validation, caching, and spawning subagents via the OpenClaw CLI. They read/write a cache under ~/.hive-task-router and reference common env vars (HIVE_MODEL_*, HIVE_VALIDATE_MODEL, etc.). Nothing in the instructions reads unrelated system secrets or posts to external URLs, but the skill will invoke model providers via OpenClaw (expected) and will spawn parallel subagents which can consume API quota and produce outbound requests to configured providers.
Install Mechanism
No install spec — instruction-only plus a shell script. No remote downloads or archive extraction are used. The script relies on the openclaw CLI being present, which is consistent with its behavior.
Credentials
The skill documents and uses HIVE_MODEL_* and related HIVE_* variables for configuration. It does not request or hard-code API keys, credentials, or other unrelated environment variables. Using OpenClaw implies provider credentials are managed elsewhere; that is proportionate to the stated multi-provider purpose.
Persistence & Privilege
always is false and the skill does not modify other skills. It does create a cache directory (~/.hive-task-router) and recommends adding env exports/aliases to shell rc files (user action). Be aware the skill may be invoked automatically by agent trigger rules and can spawn subagents; monitor concurrency limits to avoid unintended resource usage.
Assessment
This skill appears to do what it claims: route tasks and spawn OpenClaw sessions based on keywords. Before installing: (1) ensure you have and trust the OpenClaw CLI and any model provider credentials configured separately, (2) review router.sh yourself (it will run locally and write a small cache under ~/.hive-task-router), (3) set sensible limits (HIVE_MAX_CONCURRENT, HIVE_VALIDATE_MODEL) to avoid runaway parallel subagents or excessive API usage, and (4) consider disabling automatic triggers or narrowing keyword patterns if you don't want routine messages to spawn subagents automatically.Like a lobster shell, security has layers — review code before you run it.
latestvk97da6h2qn76ma7wzn882qz1d982s1kpmulti-providervk97da6h2qn76ma7wzn882qz1d982s1kpproductivityvk97da6h2qn76ma7wzn882qz1d982s1kproutingvk97da6h2qn76ma7wzn882qz1d982s1kpsubagentvk97da6h2qn76ma7wzn882qz1d982s1kptask-managementvk97da6h2qn76ma7wzn882qz1d982s1kp
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
