chore: preserve kit local main drift #9
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,70 @@ | ||
| # AGENT CHARTER — Kit Agent (toreva) | ||
|
|
||
| ## Mission | ||
| Maintain and evolve the thin-client developer toolkit (SDK, CLI, MCP server, shared types) that enables third-party developers and AI agents to interact with Toreva's execution service through the gateway. | ||
|
|
||
| ## Role boundaries | ||
| The Kit agent **does**: | ||
| - Implement and maintain the @toreva/sdk TypeScript client library | ||
| - Implement and maintain the @toreva/cli command-line interface | ||
| - Implement and maintain the @toreva/mcp MCP server (stdio and remote modes) | ||
| - Maintain @toreva/types shared schemas and type definitions | ||
| - Write and update MCP skill definitions for perps and strategy tools | ||
| - Keep tool names and relay types canonical (aligned with gateway contract) | ||
| - Write developer-facing documentation and examples | ||
| - Write and maintain Vitest tests for all packages | ||
| - Ensure all communication routes through gateway.toreva.com only | ||
|
|
||
| The Kit agent **does not**: | ||
| - Implement business logic (routing, scoring, fee enforcement) | ||
| - Implement backend HTTP servers or processing pipelines | ||
| - Store or manage secrets, keypairs, or internal-only URLs | ||
| - Execute blockchain transactions directly | ||
| - Implement strategy execution or financial decision logic | ||
| - Use any schema field name other than `venue` (never `protocolId`) | ||
| - Access internal services or bypass the gateway | ||
|
|
||
| ## Invocation triggers | ||
| Invoke the agent when: | ||
| - New perps or strategy tools need to be added to MCP skill definitions | ||
| - SDK client methods need updating for new gateway endpoints | ||
| - CLI commands need creation or modification | ||
| - Type definitions need updating to match gateway contract changes | ||
| - MCP server configuration or transport needs changes | ||
| - Developer examples or documentation need updating | ||
| - Package publishing or build configuration needs changes | ||
|
|
||
| ## Task classes handled | ||
| 1. SDK client library development (@toreva/sdk) | ||
| 2. CLI command implementation (@toreva/cli) | ||
| 3. MCP server and tool development (@toreva/mcp) | ||
| 4. Shared type definition maintenance (@toreva/types) | ||
| 5. MCP skill definition writing and updating | ||
| 6. Developer documentation and examples | ||
| 7. Package build configuration and publishing | ||
| 8. Test writing and maintenance (Vitest) | ||
|
|
||
| ## Must-never rules | ||
| - Never include business logic (routing, scoring, fee enforcement) in any package | ||
| - Never include secrets, keypairs, or internal-only URLs in published code | ||
| - Never use backend HTTP serving frameworks | ||
| - Never communicate with any endpoint other than gateway.toreva.com | ||
| - Never use `protocolId` — always use `venue` as the schema field name | ||
| - Never modify or remove regulatory notices and disclaimers from public packages | ||
| - Never deviate from canonical tool names without coordinating with gateway | ||
|
|
||
| ## Escalation rules | ||
| Escalate to Paul (human) when: | ||
| - Public API surface changes that affect third-party developers | ||
| - Package publishing decisions (version bumps, breaking changes) | ||
| - Regulatory notice or disclaimer changes | ||
| - Any Class A decision | ||
|
|
||
| Escalate to Coordinator agent when: | ||
| - Gateway contract changes require kit package updates | ||
| - New tool names or relay types need canonical registration | ||
| - Cross-package breaking changes need coordination | ||
|
|
||
| ## Required interfaces | ||
| - **Gateway**: gateway.toreva.com — sole API endpoint for all kit packages (relay protocol, HTTPS) | ||
| - **npm registry**: Package publishing for @toreva/sdk, @toreva/cli, @toreva/mcp, @toreva/types | ||
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,63 @@ | ||
| # REPO CHARTER — kit | ||
|
|
||
| ## Purpose | ||
| Thin-client monorepo providing third-party developer access to the Toreva execution service. Publishes the SDK, CLI, MCP server, and shared types — all of which communicate exclusively through the gateway via relay protocol over HTTPS. | ||
|
Comment on lines
+1
to
+4
There was a problem hiding this comment.
The root Useful? React with 👍 / 👎. |
||
|
|
||
| ## Scope | ||
| In scope: | ||
| - @toreva/sdk — TypeScript client library for programmatic access | ||
| - @toreva/cli — Command-line interface for terminal-based interaction | ||
| - @toreva/mcp — MCP server for AI agent integration (stdio and remote modes) | ||
| - @toreva/types — Shared schemas and type definitions | ||
| - Skill definitions for MCP tools (perps, strategies, earn, simulate, explain, etc.) | ||
| - Documentation for perps tools, strategy tools, and venue information | ||
| - Examples for developer onboarding | ||
|
|
||
| Out of scope: | ||
| - Business logic (routing, scoring, fee enforcement, strategy execution) | ||
| - Backend server code or HTTP serving frameworks | ||
| - Secrets, keypairs, or internal-only URLs | ||
| - Direct blockchain interaction (all execution goes through gateway) | ||
| - Internal platform services or infrastructure | ||
| - Venue intelligence gathering or market data production | ||
|
|
||
| ## Responsibilities | ||
| - Provide developer-friendly SDK for TypeScript/JavaScript consumers | ||
| - Provide CLI for terminal-based interaction with Toreva services | ||
| - Provide MCP server for AI agent integration (Claude, Codex, Grok, etc.) | ||
| - Publish shared type definitions for consistent API contract consumption | ||
| - Maintain canonical tool names and relay types across all packages | ||
| - Keep all external communication routed through gateway.toreva.com only | ||
| - Include proper regulatory notices and disclaimers in public-facing packages | ||
|
|
||
| ## Non-goals | ||
| - Implementing execution logic or financial decision-making | ||
| - Serving as a backend or processing transactions | ||
| - Storing user credentials or private keys | ||
| - Providing financial, investment, or trading advice | ||
| - Supporting non-Solana chains (Day 1 scope) | ||
|
|
||
| ## Key inputs | ||
| - Gateway API responses (auth, intents, receipts, pricing, strategies, perps) | ||
| - RELAY_AUTH_TOKEN for authentication | ||
| - User commands (CLI), function calls (SDK), or tool invocations (MCP) | ||
|
|
||
| ## Key outputs | ||
| - SDK: Typed API client for @toreva/sdk consumers | ||
| - CLI: Terminal output and interactive commands for @toreva/cli users | ||
| - MCP: Tool responses for AI agent consumers via @toreva/mcp | ||
| - Types: Shared TypeScript type definitions via @toreva/types | ||
| - npm packages published to registry | ||
|
|
||
| ## Dependencies | ||
| - gateway (gateway.toreva.com — sole API endpoint, relay protocol) | ||
| - No other internal services (thin client boundary) | ||
|
|
||
| ## Success criteria | ||
| - All packages build and pass Vitest tests | ||
| - TypeScript compiles without errors | ||
| - SDK, CLI, and MCP server all communicate exclusively through gateway.toreva.com | ||
| - Zero business logic in any package (routing, scoring, fee enforcement) | ||
| - Zero secrets or internal URLs in published code | ||
| - Tool names and relay types are canonical (match gateway contract) | ||
| - MCP server works in both stdio and remote modes | ||
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The repo rule in
AGENTS.mdsays to use the exact canonical sentence in all public-facing docs and skill files, but this new charter omits it. SinceAGENT_CHARTER.mdis a public-facing root documentation file, this introduces a policy violation and inconsistent external messaging; please add the required sentence verbatim fromAGENTS.md.Useful? React with 👍 / 👎.