Python, data analysis, charts, and document delivery; tasks can pause and resume.
Architecture / Current implementation
An always-on investment Agent
with a workspace of its own.
It does more than answer one question. It organizes research into resumable tasks, turns memory into files, and gains specialized research and delivery capabilities through Skills.
Sign in to create an independent workspace and begin research that can keep moving over time.
CORE CAPABILITIES
Six core capabilities
Profiles, portfolios, research, evidence, and Skills become a real directory tree.
Quotes, price history, news, events, and research are called through one interface, with complete evidence archived automatically.
Analysts encode evidence paths, methods, scripts, templates, and acceptance criteria into Skills.
Investment principles, formulas, historical material, and evaluations form a callable method system.
Daily briefings and reviews, weekly stock selection, and research results saved automatically.
01 / Architecture
A task-driven Agent Loop
Ingress, tasks, the Agent, capabilities, sandbox, and storage each keep a clear boundary.
REAL CLOUD SANDBOX
Not a simulated terminal, but a real cloud financial server.
Each user receives an independent Sandbox. Tasks can pause and resume while the Workspace persists. The Agent runs under a low-privilege account, with financial credentials isolated by a separate operating-system identity and file permissions.
persistent · Asia/Shanghai
Preinstalled research and delivery environment
PythonpandasNumPy
SciPystatsmodelsPyArrow
PlotlyMatplotlib
WeasyPrintopenpyxl
python-docxPDF tools
ripgrepjqNoto CJK
fin and vl sync by content digest at startup and run through controlled launchers.
02 / Workspace
Memory is a directory tree
Workspace stores organized long-term memory. Earlier conversations remain in session storage and are recalled only when needed.
workspace/ ├── AGENTS.md ├── user.yaml # identity and structured preferences ├── portfolio/ # accounts, holdings, trades, watchlists ├── stocks/ # research organized by security ├── topics/ # sectors, macro, strategies, themes ├── briefs/ # recurring briefs ├── inbox/ # user inputs, read only ├── evidence/ # data evidence, read only └── skills/ # user-installed capabilities
User model
Profile, preferences, constraints, and output habits update only through explicit requests.
Investment state
Portfolio files stay structured, with dedicated write rules loaded before updates.
Research output
Single-security work goes to stocks; cross-security and thematic work goes to topics.
Procedural memory
User Skills keep methods, scripts, and assets inside their own system.
Evidence boundary
The data plane writes to evidence; the Agent can read it but cannot overwrite the source.
Cross-session recall
When earlier work must continue, real exchanges from prior sessions are retrieved on demand.
Context freeze
The first run freezes only the map, profile, and Skill index—not the entire history.
Separate records from facts
Old sessions prove what was said then; prices, filings, and policy still require fresh data.
Files are the source of truth; context is only a cache for the current step.
03 / FIN CLI / DATA PLANE
Financial data and evidence archives
Connect current multi-asset quotes, price history and technical indicators, financial news, corporate events, and institutional research through one interface. Every call returns a concise summary and writes the complete result to evidence for later review.
Manifest drivenData-source definitions stay centralized and consistent
Evidence coveragecoverage / degraded / as_of
Deterministic failuresExit codes tell the Agent what to do next
Credential isolationSensitive credentials never enter the Sandbox
Market data quote · technical Financials financials · call Information event · news · research General websearch Complete results are written to evidence automatically, preserving source and timestamp.
04 / Skills
Professional financial workflows
Indexes stay resident while full instructions load on demand. A Skill can define evidence checklists, reasoning order, output templates, executable scripts, and acceptance rules.
listed-company-analysis
Analyze listed-company fundamentals, earnings, valuation, and competitive position.
trade-decision
Evaluate adding, trimming, chasing, stopping out, or holding a single security.
Company one-pager
A buy-side one-pager with evidence, timestamps, and visible coverage gaps.
05 / KOL Agent
An expert Avatar is more than a persona
A financial expert Avatar is a complete Skill Package. Stable methods, investment principles, calculation scripts, evaluation rules, and searchable source material determine how it researches and communicates.
kol-agent/ ├── SKILL.md # principles and research roadmap ├── references/ # principles and historical views ├── scripts/ # formulas, data retrieval, calculation, validation ├── evals/ # decision consistency and style evaluation └── .data/ # searchable source material
Master index and research roadmap
SKILL.md SKILL.md summarizes the core principles and routes tasks to the right references, scripts, data, and evaluations.
Traceable investment principles
references/ stores specific principles, historical calls, and the conditions under which they apply.
Compute and retrieve
scripts/ carries formulas and data retrieval; .data/ supports progressive search through underlying material.
Continuously evaluated
evals/ checks factual quality, decision consistency, style fidelity, and capability boundaries.
06 / Scheduled Research
Start research automatically on schedule
Set the task, run time, time zone, and model. Even after the page is closed, the system creates a new conversation at the scheduled time and runs the complete task.
Morning market briefing
Summarize overnight markets, important events, portfolio risks, and the day’s watchpoints before the open.
Daily 08:00 · Asia/Shanghai
Investment review and strategy check
Run portfolio reviews, sector tracking, or a fixed stock-selection workflow on a chosen day.
Sunday 20:00 · Selected model
Automatic post-event follow-up
Schedule one research run after earnings, a rate decision, or another known event, then end automatically.
Selected date · Local time zone