cobo-agentic-wallet-developer
Developer guide for integrating the Cobo Agentic Wallet SDK into AI agents, bots, and automation pipelines. Covers SDK installation (Python/TypeScript), `caw` CLI setup, credential management, pact-based authorization, token transfers, contract calls, framework integrations (LangChain, OpenAI Agents, CrewAI, Agno, Vercel AI SDK, Mastra), MCP server setup, and testnet testing. Use when building or debugging applications that programmatically control an agentic wallet.
What this skill does
## What You're Building
Cobo Agentic Wallet gives your AI agent a controlled on-chain runtime:
- **No private keys in the agent** — signing and key management stay outside the agent runtime
- **Pact-scoped authorization** — the agent submits a pact describing its intent; the wallet owner approves it; the agent operates only within those boundaries
- **Structured denials** — when a request is blocked, you receive a structured error with a reason and an optional suggestion for retry
- **Multi-framework** — drop the SDK as a toolkit into LangChain, OpenAI Agents SDK, CrewAI, Agno, Vercel AI SDK, Mastra, or as an MCP server
---
## Quick Setup
> Full guide: [quickstart.md](./references/quickstart.md)
```bash
# 1. Install caw CLI
curl -fsSL https://raw.githubusercontent.com/CoboGlobal/cobo-agentic-wallet/master/install.sh | bash
export PATH="$HOME/.cobo-agentic-wallet/bin:$PATH"
# 2. Onboard (provision wallet + TSS Node)
caw onboard --wait
# 3. Get credentials
caw wallet current --show-api-key
export AGENT_WALLET_API_URL=https://api.agenticwallet.cobo.com
export AGENT_WALLET_API_KEY=<your-api-key>
export AGENT_WALLET_WALLET_ID=<your-wallet-uuid>
# 4. Install SDK
pip install cobo-agentic-wallet # Python
npm install @cobo/agentic-wallet # TypeScript
```
---
## The Pact Model
Every on-chain operation (transfer, contract call, message signing) requires an active **pact**. A pact is owner-approved authorization scoped to:
- Allowed chains, tokens, and operation types
- Spending caps (per-transaction and cumulative)
- Time limit (expiry)
**Lifecycle**: `pending_approval` → `active` → `completed` / `expired` / `revoked` / `rejected`
The pact carries its own **API key** — use `pact["api_key"]` for all transactions under that pact. The pact-scoped key is automatically constrained to the approved policy; attempts to exceed it return a `PolicyDeniedError`.
```python
# Submit a pact
pact = await client.submit_pact(wallet_id=WALLET_ID, intent="...", spec={...})
pact_id = pact["pact_id"]
# Poll until active
while True:
pact = await client.get_pact(pact_id)
if pact["status"] == "active":
break
await asyncio.sleep(5)
# Use pact-scoped key for all transactions
pact_client = WalletAPIClient(base_url=API_URL, api_key=pact["api_key"])
tx = await pact_client.transfer_tokens(WALLET_ID, ...)
```
---
## SDK Reference
> Full patterns & concepts: [sdk-python.md](./references/sdk-python.md) | [sdk-typescript.md](./references/sdk-typescript.md)
>
> Authoritative API (method names, signatures): [PyPI](https://pypi.org/project/cobo-agentic-wallet/) | [npm](https://www.npmjs.com/package/@cobo/agentic-wallet)
### Core pattern
All on-chain operations follow the same two-phase flow:
1. **Submit a pact** with the onboarding key → poll until `status == "active"`
2. **Use `pact["api_key"]`** (pact-scoped key) for all transactions under that pact
```python
# Phase 1: pact management (onboarding key)
pact = await client.submit_pact(wallet_id=WALLET_ID, intent="...", spec={
"policies": [{
"name": "my-policy",
"type": "transfer",
"rules": {
"effect": "allow",
"when": {"chain_in": ["ETH"], "token_in": [{"chain_id": "ETH", "token_id": "ETH_USDC"}]},
"deny_if": {"amount_gt": "1000"},
},
}],
"completion_conditions": [{"type": "time_elapsed", "threshold": "86400"}],
})
# ... poll until pact["status"] == "active" ...
# Phase 2: transactions (pact-scoped key)
pact_client = WalletAPIClient(base_url=API_URL, api_key=pact["api_key"])
# call transfer / contract_call / sign_message via pact_client
# always set request_id for idempotency
```
The `spec.policies` structure and `completion_conditions` format are stable server-side concepts — they don't change with SDK version. Exact method names (`submit_pact`, `transfer_tokens`, etc.) may evolve; check PyPI/npm for your installed version.
---
## Framework Integrations
> Full guide: [framework-integrations.md](./references/framework-integrations.md)
| Framework | Language | Install extra | Import |
|---|---|---|---|
| LangChain | Python | `pip install cobo-agentic-wallet[langchain]` | `from cobo_agentic_wallet.integrations.langchain import CoboAgentWalletToolkit` |
| OpenAI Agents SDK | Python | `pip install cobo-agentic-wallet[openai]` | `from cobo_agentic_wallet.integrations.openai import CoboOpenAIAgentContext` |
| Agno | Python | `pip install cobo-agentic-wallet[agno]` | `from cobo_agentic_wallet.integrations.agno import CoboAgentWalletTools` |
| CrewAI | Python | `pip install cobo-agentic-wallet[crewai]` | `from cobo_agentic_wallet.integrations.crewai import CoboAgentWalletCrewAIToolkit` |
| LangChain | TypeScript | `npm install @cobo/agentic-wallet` | `import { CoboAgentWalletToolkit } from "@cobo/agentic-wallet/integrations/langchain"` |
| OpenAI Agents SDK | TypeScript | `npm install @cobo/agentic-wallet` | `import { CoboOpenAIContext } from "@cobo/agentic-wallet/integrations/openai"` |
| Vercel AI SDK | TypeScript | `npm install @cobo/agentic-wallet` | `import { createCoboTools } from "@cobo/agentic-wallet/integrations/ai-sdk"` |
| Mastra | TypeScript | `npm install @cobo/agentic-wallet` | `import { CoboMastraToolkit } from "@cobo/agentic-wallet/integrations/mastra"` |
| MCP | Python | `pip install cobo-agentic-wallet[mcp]` | `python -m cobo_agentic_wallet.mcp` |
Use `include_tools` / `exclude_tools` to narrow the tool surface exposed to the agent.
---
## MCP Server
> Full guide: [mcp-server.md](./references/mcp-server.md)
```bash
# Start MCP server (stdio transport — for Claude Desktop / Cursor)
AGENT_WALLET_API_KEY=<key> AGENT_WALLET_WALLET_ID=<uuid> \
python -m cobo_agentic_wallet.mcp
# Or via uvx (no install needed)
uvx cobo-agentic-wallet[mcp]
```
Add to `claude_desktop_config.json`:
```json
{
"mcpServers": {
"cobo-wallet": {
"command": "python",
"args": ["-m", "cobo_agentic_wallet.mcp"],
"env": {
"AGENT_WALLET_API_KEY": "<key>",
"AGENT_WALLET_WALLET_ID": "<uuid>",
"AGENT_WALLET_API_URL": "https://api.agenticwallet.cobo.com"
}
}
}
}
```
---
## Testing with Testnet
> Full guide: [testing.md](./references/testing.md)
```bash
# Get a Sepolia address
caw address list
# Fund it from the faucet
caw faucet deposit --token-id SETH --address <your-seth-address>
caw faucet deposit --token-id SETH_USDC --address <your-seth-address>
# Check balance
caw wallet balance
```
For testnet, use `chain_id=SETH` (Ethereum Sepolia) and `token_id=SETH` / `SETH_USDC`.
Change the API URL to the sandbox environment if you have a sandbox account:
```bash
export AGENT_WALLET_API_URL=https://api.sandbox.agenticwallet.cobo.com
```
---
## caw CLI for Developers
```bash
# Check wallet + credentials + active pacts
caw status
# Show credentials (api_key, api_url, wallet_uuid)
caw wallet current --show-api-key
# List all local profiles
caw wallet list
# Switch active profile (if running multiple wallets)
caw wallet current --wallet-id <wallet-uuid>
# Look up supported chains and tokens
caw meta chains
caw meta tokens --chain-ids ETH
caw meta tokens --token-ids ETH_USDC,SETH_USDC
# Get schema for any command (exact flags + types)
caw schema tx transfer
caw schema pact submit
# Submit a pact (required flags + optional metadata flags)
# --name and --recipe-slugs are optional; all other flags below are required
caw pact submit \
--intent "Transfer 100 USDC to 0xABC on Base" \
--execution-plan "# Summary\nTransfer 100 USDC.\n# Operations\n- Transfer 100 USDC to 0xABC on Base\n# Risk Controls\n- Per-tx cap: $101" \
--policies '[{"name":"usdc-transfer","type":"transfer","rules":{"effect":"allow","when":{"chain_in":["BASE_ETH"],"token_in":[{"chain_id":"BASE_ETH","token_id":"BASE_USDC"}]},"deny_if":{"amount_usd_gt":"101"}}}]' \
--completion-conditions '[{"type":"tx_count","threshold":"1"}]' \
--name "USDC transfer to 0xABC" \
--recipe-slugs "uniswRelated in Backend & APIs
jfrog
IncludedInteract with the JFrog Platform via the JFrog CLI and REST/GraphQL APIs. Use this skill when the user wants to manage Artifactory repositories, upload or download artifacts, manage builds, configure permissions, manage users and groups, work with access tokens, configure JFrog CLI servers, search artifacts, manage properties, set up replication, manage JFrog Projects, run security audits or scans, look up CVE details, query exposures scan results from JFrog Advanced Security, manage release bundles and lifecycle operations, aggregate or export platform data, or perform any JFrog Platform administration task. Also use when the user mentions jf, jfrog, artifactory, xray, distribution, evidence, apptrust, onemodel, graphql, workers, mission control, curation, advanced security, exposures, or any JFrog product name.
cupynumeric-migration-readiness
IncludedPre-migration readiness assessor for porting NumPy to cuPyNumeric. Use BEFORE substantial porting work begins when the user asks whether code will scale on GPU, whether they should migrate to cuPyNumeric, which NumPy patterns transfer cleanly, what must be refactored before porting, or mentions pre-port assessment, scaling analysis, or refactor planning. Inspect the user's source code, look up NumPy usage, cross-reference the cuPyNumeric API support manifest, and distinguish distributed-scaling-friendly patterns from blockers such as unsupported APIs, scalar synchronization, host round-trips, Python/object-heavy control flow, shape/data-dependent branching, and in-place mutation hazards. Produce a verdict of READY, LIGHT REFACTOR, SIGNIFICANT REFACTOR, or NOT RECOMMENDED, with concrete refactor pointers.
alibabacloud-data-agent-skill
IncludedInvoke Alibaba Cloud Apsara Data Agent for Analytics via CLI to perform natural language-driven data analysis on enterprise databases. Data Agent for Analytics is an intelligent data analysis agent developed by Alibaba Cloud Database team for enterprise users. It automatically completes requirement analysis, data understanding, analysis insights, and report generation based on natural language descriptions. This tool supports: discovering data resources (instances/databases/tables) managed in DMS, initiating query or deep analysis sessions, real-time progress tracking, and retrieving analysis conclusions and generated reports. Use this Skill when users need to query databases, analyze data trends, generate data reports, ask questions in natural language, or mention "Data Agent", "data analysis", "database query", "SQL analysis", "data insights".
token-optimizer
IncludedReduce OpenClaw token usage and API costs through smart model routing, heartbeat optimization, budget tracking, and native 2026.2.15 features (session pruning, bootstrap size limits, cache TTL alignment). Use when token costs are high, API rate limits are being hit, or hosting multiple agents at scale. The 4 executable scripts (context_optimizer, model_router, heartbeat_optimizer, token_tracker) are local-only — no network requests, no subprocess calls, no system modifications. Reference files (PROVIDERS.md, config-patches.json) document optional multi-provider strategies that require external API keys and network access if you choose to use them. See SECURITY.md for full breakdown.
resend-cli
IncludedUse this skill when the task is specifically about operating Resend from an AI agent, terminal session, or CI job via the official resend CLI: installing/authenticating the CLI, sending/listing/updating/cancelling emails, batch sends, domains and DNS, webhooks and local listeners, inbound receiving, contacts, topics, segments, broadcasts, templates, API keys, profiles, or debugging Resend CLI/API failures. Trigger on mentions of Resend CLI, `resend`, `resend doctor`, `resend emails send`, `resend domains`, `resend webhooks listen`, `resend emails receiving`, or agent-friendly terminal automation.
alibabacloud-odps-maxframe-coding
IncludedUse this skill for MaxFrame SDK development and documentation navigation on Alibaba Cloud MaxCompute (ODPS). Helps answer MaxFrame API, concept, official example, and supported pandas API questions; create data processing programs; read/write MaxCompute tables; debug jobs (remote or local); and build custom DPE runtime images. Trigger when users mention MaxFrame, MaxCompute with MaxFrame, ODPS table processing, DPE runtime, MaxFrame docs/examples, DataFrame/Tensor operations, or GPU runtime setup. Works for both English and Chinese queries about Alibaba Cloud data processing with MaxFrame.