trace-claude-code
Automatically trace Claude Code conversations to Braintrust for observability. Captures sessions, conversation turns, and tool calls as hierarchical traces.
Graceful Degradation with Helpful Messages
$ npx -y skills add parcadei/Continuous-Claude-v3 --skill graceful-degradation --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/graceful-degradationContext preview
The summary Claude sees to decide when to auto-load this skill.
Graceful Degradation with Helpful Messages
name: graceful-degradation description: Graceful Degradation with Helpful Messages user-invocable: false
When optional services are unavailable, degrade gracefully with actionable fallback messages.
Check availability at the start, cache the result, and provide helpful messages that explain what's missing and how to fix it.
let lmstudioAvailable: boolean | null = null;
let lastCheck = 0;
const CACHE_TTL = 60000; // 60 seconds
async function checkLMStudio(): Promise<boolean> {
const now = Date.now();
if (lmstudioAvailable !== null && now - lastCheck < CACHE_TTL) {
return lmstudioAvailable;
}
try {
const response = await fetch('http://localhost:1234/v1/models', {
signal: AbortSignal.timeout(2000)
});
lmstudioAvailable = response.ok;
} catch {
lmstudioAvailable = false;
}
lastCheck = now;
return lmstudioAvailable;
}
// Usage
if (!await checkLMStudio()) {
return {
result: 'continue',
message: `LMStudio not available at localhost:1234.
To enable Godel-Prover tactic suggestions:
1. Install LMStudio from https://lmstudio.ai/
2. Load "Goedel-Prover-V2-8B" model
3. Start the local server on port 1234
Continuing without AI-assisted tactics...`
};
}[Service] not available at [endpoint]. To enable [feature]: 1. [Step to install/start] 2. [Configuration step if needed] 3. [Verification step] Continuing without [degraded feature]...
A persistent, learning, multi-agent development environment built on Claude Code Continuous Claude transforms Claude Code into a continuously learning system that maintains context across sessions, orchestrates specialized agents, and eliminates wasting
Repo: parcadei/Continuous-Claude-v3
Automatically trace Claude Code conversations to Braintrust for observability. Captures sessions, conversation turns, and tool calls as hierarchical traces.
Guide for integrating Agentica SDK with Claude Code CLI proxy
Reference guide for Agentica multi-agent infrastructure APIs