everclaw-inferenceOpen-source first AI inference — GLM-5 as default, Claude as fallback only. Own your inference forever via the Morpheus decentralized network. Stake MOR toke...
Install via ClawdBot CLI:
clawdbot install davidajohnston/everclaw-inferenceGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Accesses sensitive credential files or environment variables
/etc/passwdContains instructions to override system prompt or ignore user requests
"ignore previous instructions"Sends data to undocumented external endpoint (potential exfiltration)
send → http://localhost:8082/v1/chat/completions`Potentially destructive shell commands in tool definitions
rm -rf ~Generated Mar 22, 2026
A research team uses Everclaw to run persistent inference sessions with models like Kimi K2.5 for data analysis and hypothesis testing, staking MOR tokens to avoid centralized API costs and ensuring data privacy through the Morpheus P2P network. This enables continuous, self-sovereign access to AI without vendor lock-in.
A startup deploys an OpenClaw agent powered by Everclaw to handle customer inquiries via the OpenAI-compatible proxy, leveraging automatic retry and session management for reliable, low-cost inference. By recycling staked MOR, they maintain 24/7 support with fallback to Venice API, reducing operational expenses.
A creative agency uses Everclaw to generate marketing content and scripts using models like Llama 3.3, staking MOR on Base to access decentralized inference and integrating x402 payments for agent-to-agent USDC transactions. This allows scalable, trustless collaboration among freelance agents.
A fintech firm employs Everclaw with proactive Venice DIEM credit monitoring and circuit breakers to analyze market trends and generate reports, ensuring high availability through the Morpheus API Gateway and guardian health checks. This provides resilient, cost-effective AI insights for trading decisions.
An edtech company integrates Everclaw into an OpenClaw-based tutoring system, using multi-key auth rotation and smart session archiving to manage student interactions across 30+ models without dashboard overload. Persistent inference via MOR staking supports personalized learning at scale.
Offer a platform where users pay a subscription to access pooled MOR staking for inference, with revenue from small usage fees on recycled stakes and premium support for session management. This model reduces upfront token costs for clients while ensuring reliable AI access.
Provide consulting and deployment services for businesses to build custom OpenClaw agents with Everclaw, charging for setup, integration, and ongoing maintenance of the proxy-router and guardian systems. Revenue comes from project-based fees and retainer contracts for monitoring.
Create a marketplace connecting model providers with consumers via the Morpheus network, earning commissions on MOR swaps and session fees while offering tools like ERC-8004 registry discovery. This facilitates trustless agent interactions and expands the ecosystem.
💬 Integration Tip
Start by testing with the included Morpheus API Gateway bootstrap for zero-config setup, then gradually integrate MOR staking for persistent inference to optimize costs and reliability.
Scored May 17, 2026
Accesses system directories or attempts privilege escalation
/etc/hostsCalls external URL not in known-safe list
https://everclaw.comUses known external API (expected, informational)
api.anthropic.comAI Analysis
The skill's external API usage (api.mor.org, provider.mor.org, api.venice.ai) is consistent with its stated purpose of decentralized AI inference, and the credential handling via macOS Keychain/1Password is a secure pattern. The 'ignore previous instructions' signal is likely a false positive from a user prompt example, not a system override, and the localhost data sink is for its own proxy service, not exfiltration.
Audited Apr 17, 2026 · audit v1.0
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
通过网页抓取获取 GitHub 按日/周/月增长的热门仓库。当用户询问 GitHub 趋势、热门项目、本周热点或「什么在 GitHub 上 trending」时使用。可输出列表或 JSON,无需 API Key。
GitHub 操作技能 - 创建仓库、推送代码、管理 Release。全自动,无需用户干预。
Query builder reputation data via Talent Protocol API. Get Builder Rank, verify humans, resolve identities (Twitter/Farcaster/GitHub/wallet), search by location/country, get credentials, and enrich with GitHub data.
Review GitHub repositories, group issues and pull requests by urgency, and take safe triage actions (add labels, comment, propose closure) after confirmation...
Semantic git history search and code archaeology. Use when asked why code exists, who owns a file, what introduced a regression, what changed in a commit ran...