skills.rest
diazMelgarejo
diazMelgarejoCOMMUNITY
Explore all skills by this author →
Website𝕏@Melgarejo_yDiazGitHub

glm52-fallback

Configures BigModel GLM-5.2 as a system-wide LLM fallback using environment-variable API key authentication.

Updated Mar 22, 2026
Search Tags:#bash#api key#openclaw#provider failover#llm fallback#glm-5.2#bigmodel
One-click install
npx skills add https://github.com/diazMelgarejo/orama-system --skill glm52-fallback-diazmelgarejo
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: glm52-fallback
Source: https://github.com/diazMelgarejo/orama-system/tree/main/bin/orama-system/skills/glm52-fallback
Command: npx skills add https://github.com/diazMelgarejo/orama-system --skill glm52-fallback-diazmelgarejo

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When ClinePass (Cline Credits) is unavailable or exhausted, agents lose their primary LLM provider. This Skill sets up BigModel GLM-5.2 as a system-wide fallback so all agents keep working without manual reconfiguration or hardcoded credentials. ## Core Features & Use Cases - Automated Fallback Setup: A bash script stores the API key in a mode-600 secrets file, creates an operator-local env bundle, and tests the BigModel endpoint connection. - Env-Var-Only Authentication: Reads the key from $GLM52_API_KEY or $OPENCLAW_MODELS_PROVIDERS_BIGMODEL_APIKEY and fails fast with a clear error if unset, making it safe for unattended CI runs. - Defined Failover Chain: Documents a four-tier fallback order (ClinePass, BigModel GLM-5.2, OpenRouter free tier, local Ollama) with timeout, retry, and exit-code policies. - Use Case: An operator whose Cline Credits run out mid-session exports GLM52_API_KEY and runs the setup script, restoring GLM-5.2 access for every agent on the next startup. ## Quick Start Set the GLM52_API_KEY environment variable and ask the agent to run the GLM-5.2 fallback setup script, then verify the connection reports healthy.

Frequently Asked Questions about glm52-fallback

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I set up GLM-5.2 as a fallback LLM provider?▼

Export the GLM52_API_KEY environment variable, then run the setup-glm52.sh script. It stores the key in a mode-600 secrets file, creates an env bundle under ~/.openclaw, and tests the BigModel endpoint connection.

What is the fallback order when ClinePass credits run out?▼

The chain is: ClinePass first, then BigModel GLM-5.2 via open.bigmodel.cn, then OpenRouter free tier, and finally local Ollama offline. This skill configures the second tier.

Can I run the GLM-5.2 setup script in CI without interaction?▼

Yes. The script fails fast with a clear error if GLM52_API_KEY is unset instead of prompting, so it is safe for unattended or CI execution. It never hardcodes or prints the credential value.

Does the setup script modify my shell profile automatically?▼

No. Shell profile persistence is opt-in via GLM52_PERSIST_SHELL_PROFILE=1. The runtime start.sh already sources the GLM52 env bundle, so profile changes are only needed for use outside orama runs.

Why does the GLM-5.2 connection test fail after setup?▼

Failures usually mean an invalid or unfunded BigModel API key, or network issues reaching open.bigmodel.cn. Exit codes distinguish causes: 7 for connection failure, 28 for timeout, and 52 for empty reply.

Related Skills

Explore More Software Engineering Skills→
SynthFlowAISynthFlowAI

setup-api-key

Configure a Synthflow API key in .env and validate it with curl.

Official
Intermediate
Renzo-TognellaRenzo-Tognella

Multi-Provider LLM

Configure and switch between OpenAI, Z.ai, OpenRouter, Gemini, and local fallback providers.

Community
Advanced
bereniketechbereniketech

multi-model-model-agnostic-platforms

Route tasks across multiple AI models with cost-aware fallback strategies.

Community
Advanced
ollieb89ollieb89

freeride

Manage free OpenRouter AI models for OpenClaw with automatic ranking and fallbacks.

Community
Intermediate