emblem-ai-agent-wallet
Connect to EmblemVault and manage crypto wallets via EmblemAI. Supports Solana, Ethereum, Base, BSC, Polygon, Hedera, and Bitcoin. Use when the user wants to trade crypto, check balances, swap tokens, interact with blockchain wallets, or create and use profile-scoped agent wallets.
What this skill does
# Emblem Agent Wallet Connect to **EmblemAI** -- EmblemVault's autonomous crypto AI with 250+ trading tools across 7 blockchains. Browser auth, streaming responses, profile-scoped local state, plugin system, and zero-config agent mode. **Requires the CLI**: `npm install -g @emblemvault/agentwallet` --- ## Quick Start -- How to Use This Skill **Step 1: Install the CLI** ```bash npm install -g @emblemvault/agentwallet ``` This provides a single command: `emblemai` **Step 2: Use it** When this skill loads, you can ask EmblemAI anything about crypto: - "What are my wallet addresses?" - "Show my balances across all chains" - "What's trending on Solana?" - "Swap $20 of SOL to USDC" - "Send 0.1 ETH to 0x..." For zero-config agent provisioning, a profile can create and persist its own wallet with a single command: ```bash emblemai --agent --profile hustle -m "What are my wallet addresses?" ``` **To invoke this skill, say things like:** - "Use my Emblem wallet to check balances" - "Ask EmblemAI what tokens I have" - "Connect to EmblemVault" - "Check my crypto portfolio" - "Create or use my agent wallet profile and show my addresses" All requests are routed through `emblemai` under the hood. If more than one profile exists, every agent-mode invocation must include `--profile <name>`. --- ## Prerequisites - **Node.js** >= 20.18.0 - **Terminal** with 256-color support (iTerm2, Kitty, Windows Terminal, or any xterm-compatible terminal) - **Optional**: [glow](https://github.com/charmbracelet/glow) for rich markdown rendering (`brew install glow` on macOS) ## Installation ### From npm (Recommended) ```bash npm install -g @emblemvault/agentwallet ``` ### From source ```bash git clone https://github.com/EmblemCompany/EmblemAi-cli.git cd EmblemAi-cli npm install npm link # makes `emblemai` available globally ``` ## First Run 1. Install: `npm install -g @emblemvault/agentwallet` 2. Create or pick a profile: `emblemai profile create hustle` 3. Run either `emblemai --profile hustle` or `emblemai --agent --profile hustle -m "What are my wallet addresses?"` 4. Check `/plugins` to see which plugins loaded 5. Type `/help` to see all commands 6. Back up profile auth immediately after first wallet creation --- ## Authentication EmblemAI v3 supports two authentication methods with profile-scoped local state: **browser auth** for interactive use and **password auth** for agent/scripted use. ### Profile Rules Profiles are the canonical multi-agent isolation mechanism. - `emblemai profile list` - `emblemai profile create <name>` - `emblemai profile use <name>` - `emblemai profile inspect [name]` - `emblemai profile delete <name>` - `emblemai --profile <name> ...` **Fail closed rule:** if more than one profile exists in `~/.emblemai`, every `--agent` invocation must include `--profile <name>`. Agent mode never guesses which wallet identity to use. Using separate `HOME` directories is optional isolation, not the primary pattern. Prefer profiles first. ### Browser Auth (Interactive Mode) When you run `emblemai --profile <name>` without `-p`, the CLI: 1. Checks the selected profile for a saved session 2. If a valid (non-expired) session exists, restores it instantly -- no login needed 3. If no session, starts a local server on `127.0.0.1:18247` and opens your browser 4. You authenticate via the EmblemVault auth modal in the browser 5. The session JWT is captured, saved to that profile on disk, and the CLI proceeds 6. If the browser can't open, the URL is printed for manual copy-paste 7. If authentication times out (5 minutes), falls back to a password prompt for the same profile ### Password Auth (Agent Mode) **Login and signup are the same action.** The first use of a password creates a vault; subsequent uses of the same password return the same vault. Different passwords produce different wallets, while profiles isolate the local session and stored credentials around those wallets. In agent mode, if no password is provided for the selected profile, a secure random password is auto-generated and stored encrypted via dotenvx under `~/.emblemai/profiles/<name>/`. Agent mode works out of the box with no manual setup. ### What Happens on Authentication 1. The selected profile is resolved 2. Browser auth: session JWT is received from browser and hydrated into the SDK Password auth: password is sent to `EmblemAuthSDK.authenticatePassword()` and can be stored in the selected profile for reuse 3. A deterministic vault is derived -- same credentials always yield the same vault 4. The session provides wallet addresses across multiple chains: Solana, Ethereum, Base, BSC, Polygon, Hedera, Bitcoin 5. The EmblemAI client is initialized with the profile's session ### Credential Discovery Before making requests for a profile, locate the password using this priority: | Method | How to use | Priority | |--------|-----------|----------| | CLI argument | `emblemai --profile <name> -p "your-password"` | 1 (highest, stored encrypted) | | Environment variable | `EMBLEM_PASSWORD="your-password" emblemai --profile <name>` | 2 (not stored) | | Encrypted credential | dotenvx-encrypted `~/.emblemai/profiles/<name>/.env` | 3 | | Auto-generate (agent mode) | `emblemai --agent --profile <name> ...` on first run | 4 | | Interactive prompt | Fallback when browser auth fails | 5 (lowest) | Interactive mode resolves auth for the selected profile in this order: saved session, stored password, browser auth modal, then terminal password prompt. If no credentials are found for the selected profile, agent mode can auto-generate and store them automatically. Ask the user for a password only when they explicitly want to connect an existing password-derived wallet: > "I can create a fresh EmblemVault wallet for this profile automatically, or connect an existing wallet if you enter the original password locally in the CLI. > > **Note:** If this is your first time, entering a new password will create a new wallet. If you've used this before, use the same password to access your existing wallet. > > Would you like to provide a password for this profile in the terminal?" - Password must be 16+ characters - No recovery if lost (treat it like a private key) --- ## Execution Notes **Allow sufficient time.** EmblemAI queries may take up to 2 minutes for complex operations (trading, cross-chain lookups). The CLI outputs progress dots every 5 seconds to indicate it's working. **Present EmblemAI's response clearly.** Display the response from EmblemAI to the user in a markdown codeblock: ```markdown **EmblemAI Response:** \`\`\` [response from EmblemAI] \`\`\` ``` --- ## Usage ### Agent Mode (For AI Agents -- Single Shot) Use `--agent` mode for programmatic, single-message queries with profile-scoped auth: ```bash # Zero-config -- auto-generates a wallet for the profile on first run emblemai --agent --profile hustle -m "What are my wallet addresses?" # Explicit password for a profile emblemai --agent --profile hustle -p "$PASSWORD" -m "Show my balances" # Pipe output to other tools emblemai -a --profile treasury -m "What is my SOL balance?" | jq . # Use in scripts ADDRESSES=$(emblemai -a --profile treasury -m "List my addresses as JSON") ``` Any system that can shell out to a CLI can give its agents a wallet: ```bash # OpenClaw, CrewAI, AutoGPT, or any agent framework emblemai --agent --profile ops -m "Send 0.1 SOL to <address>" emblemai --agent --profile treasury -m "Swap 100 USDC to ETH on Base" emblemai --agent --profile research -m "What tokens do I hold across all chains?" ``` Profiles are the primary way to isolate agent wallets. To give multiple agents separate wallets, use different profiles: ```bash emblemai --agent --profile agent-alice -m "My addresses?" emblemai --agent --profile agent-bob -m "My addresses?" ``` Agent mode always uses password auth (never browser auth), retains conversation history between calls for the selected profile, and supports the
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.