repositories
loading repo index
repositories
loading repo index
repository
loading code, commits, and activity
public Clawd ADK gateway launch mirror
stars
latest
clone command
git clone gitlawb://did:key:z6Mkq5mY...iFZ5/my-project-publ...git clone gitlawb://did:key:z6Mkq5mY.../my-project-publ...2fa351d6docs: add automaton and perps launch sources16d ago| #1 | { |
| #2 | "skillId": "pump-ai-agents", |
| #3 | "name": "pump-ai-agents", |
| #4 | "description": "AI agent integration layer for the Pump SDK — agent instruction files, .well-known discovery, LLM context documents, 15+ skill files, MCP server prompts, and terminal management rules for GitHub Copilot, Claude, and Gemini.", |
| #5 | "category": "pump-protocol", |
| #6 | "path": "pump-ai-agents/SKILL.md", |
| #7 | "url": "https://x402.wtf/api/skills/pump-ai-agents", |
| #8 | "tags": [ |
| #9 | "pump", |
| #10 | "solana", |
| #11 | "ai", |
| #12 | "agents" |
| #13 | ], |
| #14 | "requiredEnv": [], |
| #15 | "homepage": "https://github.com/nirholas/pump-fun-sdk", |
| #16 | "attestation": { |
| #17 | "status": "pending", |
| #18 | "isFormallyVerified": false, |
| #19 | "attestationPda": null, |
| #20 | "verificationTimestamp": null |
| #21 | }, |
| #22 | "markdown": "---\nname: pump-ai-agents\ndescription: \"AI agent integration layer for the Pump SDK — agent instruction files, .well-known discovery, LLM context documents, 15+ skill files, MCP server prompts, and terminal management rules for GitHub Copilot, Claude, and Gemini.\"\nmetadata:\n openclaw:\n homepage: https://github.com/nirholas/pump-fun-sdk\n---\n\n# AI Agent Integration — Agent Scaffolding & Discovery\n\nConfigure AI agents (GitHub Copilot, Claude, Gemini) to work effectively with the Pump SDK through instruction files, skill registries, and MCP server integration.\n\n## Agent Instruction Files\n\n| File | Agent | Purpose |\n|------|-------|---------|\n| `AGENTS.md` | Universal | Project overview, architecture, security rules |\n| `CLAUDE.md` | Claude Code | Key patterns, critical rules |\n| `COPILOT.md` | GitHub Copilot | Instructions pointer |\n| `GEMINI.md` | Google Gemini | Instructions pointer |\n| `.github/copilot-instructions.md` | Copilot Chat | SDK patterns, security rules |\n\n## .well-known Discovery Files\n\n| File | Purpose |\n|------|---------|\n| `.well-known/ai-plugin.json` | AI plugin manifest |\n| `.well-known/agent.json` | Agent capabilities and config |\n| `.well-known/skills.json` | Skills registry (15+ skills) |\n| `.well-known/security.txt` | Security contact information |\n\n## LLM Context Documents\n\n| File | Size | Purpose |\n|------|------|---------|\n| `llms.txt` | Short | Quick reference for LLMs |\n| `llms-full.txt` | Long | Comprehensive LLM context |\n\n## Skills Registry Schema\n\n```json\n{\n \"schema_version\": \"v1\",\n \"name\": \"pump-fun-sdk\",\n \"skills\": [\n {\n \"id\": \"pump-sdk-core\",\n \"name\": \"Pump SDK Core\",\n \"description\": \"...\",\n \"path\": \".github/skills/pump-sdk-core.skill.md\",\n \"tags\": [\"typescript\", \"solana\", \"sdk\"]\n }\n ]\n}\n```\n\n## Terminal Management (Agent Rules)\n\n- **Always use background terminals** (`isBackground: true`) for every command\n- **Always kill the terminal** after completion — never leave terminals open\n- Do not reuse foreground shell sessions\n- If a terminal appears unresponsive, kill it and create a new one\n\n## Security Invariants for Agents\n\n- ONLY official Solana Labs crypto: `solana-sdk`, `@solana/web3.js`, `solana-keygen`\n- Zeroize all key material after use\n- File permissions `0600` for keypairs\n- No network calls for key generation\n- All amounts use `BN` (bn.js) — never JavaScript `number` for financial math\n- `createInstruction` (v1) is deprecated — use `createV2Instruction`\n\n## MCP Server Integration\n\n```json\n{\n \"mcpServers\": {\n \"solana-wallet\": {\n \"command\": \"npx\",\n \"args\": [\"-y\", \"@pump-fun/mcp-server\"],\n \"transportType\": \"stdio\"\n }\n }\n}\n```\n\n## Patterns to Follow\n\n- Keep agent instruction files concise and actionable\n- Point agents to specific skill files for domain knowledge\n- Update `skills.json` when adding or modifying skill files\n- Maintain consistency between `AGENTS.md`, `CLAUDE.md`, and `COPILOT.md`\n\n## Common Pitfalls\n\n- Agent instruction files that are too long get truncated or ignored\n- Missing skills from `skills.json` means agents won't discover them\n- Terminal management rules are critical in Codespaces — stale terminals block operations\n- MCP server must be registered in the agent's tool configuration\n\n" |
| #23 | } |
| #24 |