The README had grown to ~600 lines, mostly MCP and HTTP API reference. It now keeps the pitch, installation, quick start, a short TUI section, the command table and an index of docs/. The reference material moved to docs/ (tui, agents, mcp, http-api, library, quotes, configuration), and the MCP docs test now checks docs/mcp.md. The TUI screenshots and the CLI demo gif were outdated and are removed; img/ is excluded from the published crate. Em-dashes are gone from the README, the new docs and the crate description.
1.5 KiB
AI and agent integration
portfolio_rs is designed to be operated by coding agents and LLMs, locally:
-
Structured output:
context,review, andsimulatesupport--format json(camelCase) for scripts and agents, and Markdown for humans and LLM prompts.portfolio_rs context positions.json --format json portfolio_rs review positions.json --policy portfolio/policy.toml --format json -
Machine-readable policy:
portfolio/policy.tomlencodes your goals, risk profile, target allocations, and constraints. Strategy templates:balanced-growth,capital-preservation,aggressive-growth,custom. -
Portable agent skill: install the built-in
portfolio-rsskill into any agent harness that supports skill files (opencode, Claude Code, Cursor, ...):portfolio_rs agent skill export ~/.config/opencode/skills -
Workspace instructions:
portfolio_rs agent initcreatesAGENTS.mdandCLAUDE.mdwith local paths and safety rules (no trades, no broker interaction, private data stays local). -
Durable memory: decisions and reports are plain Markdown files in your workspace: reviewable in six months, greppable forever. Use
--dry-runto preview any file mutation. -
MCP server:
portfolio_rs mcpserves your portfolio read-only to MCP clients over stdio (see MCP server). -
HTTP API:
portfolio_rs apiserves the same data to local scripts and GUIs (see HTTP API).