quantdinger-agent-workflow

A set of instructions for coding agents working in the QuantDinger repository, especially on its Python API, trading strategies, deployment files, and agent documentation. It defines the repository's documentation, command, and optional web or tool interfaces.

In plain words
What is it for?
Use it when editing QuantDinger backend code, trading or backtesting logic, Docker and scripts, or agent-facing documentation.
Why use it?
It helps changes follow the repository's existing contracts and safety boundaries. It points agents to the authoritative design, API, and quickstart documents before implementation.

Skill for Claude CodeCodexCursor

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/openbyteinc/quantdinger/quantdinger-agent-workflow
Any agent
npx skills add OpenByteInc/QuantDinger --skill quantdinger-agent-workflow
Clone the repo
git clone --depth 1 https://github.com/OpenByteInc/QuantDinger

Made for: Claude Code, Codex, Cursor.

Per session 47 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,103 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00047 $0.01103
Opus 5 $0.00023 $0.00551
Sonnet 5 $0.00009 $0.00221
Haiku 4.5 $0.00005 $0.00110

Measured yesterday against content hash 33614625a305, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

quantdinger-agent-workflow scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.cursor/skills/quantdinger-agent-workflow/SKILL.md · 80 lines

How it starts

The opening of the file, as written. The whole thing — 80 lines — stays where its author put it; the contents beside it link to each section on GitHub.

QuantDinger agent workflow

When this applies

Use this skill whenever you change code or docs under this repository as a coding agent (Cursor, Claude Code, Codex, or similar), especially:

  • backend_api_python/ (Flask API, services, routes)
  • Strategy / backtest / trading-adjacent logic
  • docker-compose.yml, scripts/, env.example
  • docs/agent/ (keep English only)

Read first

  1. docs/agent/AGENT_ENVIRONMENT_DESIGN.md - SSOT for three layers: documentation contract -> command contract -> optional HTTP/MCP.
  2. docs/agent/AI_INTEGRATION_DESIGN.md - How external AI agents consume QuantDinger as a product (Agent Gateway, scopes, MCP, trading safety). Read this before adding any new endpoint or tool that an AI agent might call.
  3. docs/agent/AGENT_QUICKSTART.md - Operator/integrator walkthrough; mirrors the implemented /api/agent/v1 surface.
  4. docs/agent/agent-openapi.json - Machine-readable contract; update it whenever you add or change an /api/agent/v1/... route.
  5. docs/agent/README.md - Index of agent-facing docs.

Implemented surface (truth)

The Agent Gateway is mounted at /api/agent/v1 by app/routes/agent_v1/.

  • Auth: app/utils/agent_auth.py (@agent_required(scope=...)). Tokens are hashed at rest in qd_agent_tokens; never log or persist the raw token.
  • Async jobs: app/utils/agent_jobs.py writes to qd_agent_jobs; backtests and experiment pipelines submit here and clients poll /jobs/{id} or subscribe to GET /jobs/{id}/stream (SSE: snapshot / progress / ping / result). Long-running runners can opt in by adding a second positional on_progress parameter. submit_job will detect it and pipe events to live SSE subscribers AND persist the latest snapshot.
  • Audit: every call (success and denial) is appended to qd_agent_audit.
  • Trading: quick_trade.py enforces paper-only by default; live execution requires both paper_only=false on the token AND env AGENT_LIVE_TRADING_ENABLED=true. Do not weaken this without explicit ask.
  • MCP: mcp_server/ is a thin Python wrapper over R + W + B endpoints (no trading), with three transports selected by QUANTDINGER_MCP_TRANSPORT: stdio (default, desktop IDEs), sse, and streamable-http (cloud agents / remote IDEs; also bind QUANTDINGER_MCP_HOST / QUANTDINGER_MCP_PORT). Add new tools there only after exposing the underlying capability via REST.
  • Admin UI: the Vue project at QuantDinger-Vue-src/ ships Profile -> My Agent Token for every logged-in user (src/views/profile/components/ProfileAgentTokens.vue, API /api/agent/v1/me/tokens). Admins retain /agent-tokens for audit. API client lives in src/api/agent.js.

Read the full file on GitHub · 80 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 80 lines · 47 tokens per session scan A 33614625a305

Subscribe to this mod's changes

quantdinger-agent-workflow is a skill published in the GitHub repository OpenByteInc/QuantDinger (11,264 stars, last pushed today), licensed Apache-2.0. It adds 47 tokens to every session and 1,103 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

options-trading

Use before researching, selecting, opening, modifying, or closing any option position, including single-leg options, vertical spreads, iron condors, butterflies, straddles, strangles, calendars, and other multi-leg structures. Also use when a broad trading mandate leads you to consider options even if the user did not…

Lumiwealth/lumibot · 71 tokens

stock-trading

Use before researching, selecting, opening, modifying, or closing a stock or ETF position, including discretionary investing, rotation, breakout, momentum, mean-reversion, opening-range breakout, and VWAP trading. Also use when a broad mandate leads you to consider stocks or ETFs even if the user did not name an asset…

Lumiwealth/lumibot · 70 tokens

google-agents-cli-deploy-readiness-aisp

This skill should be used when the user wants to "deploy with human approval", "verify a deployment approval", "resolve the effective agents-cli deploy plan", or needs a forced-blocking gate before agents-cli deploy. Covers resolving CLI flags + agents-cli-manifest.yaml into a plan, presenting that plan at…

babyblueviper1/invinoveritas · 151 tokens

invinoveritas

Pay-per-call agent tools over Bitcoin Lightning / USDC (x402), exposed as a remote MCP server. Use BEFORE any irreversible or consequential action (a trade, a destructive command, shipping code, spending funds) to get a capital-scale-aware governance review; for facts-only crypto market intelligence (macro risk…

babyblueviper1/invinoveritas · 175 tokens

verification-handshake

The agent-to-agent trust handshake. Use whenever you are about to ACT ON another agent's output, claim, or deliverable that you cannot independently verify, AND whenever you PRODUCE output that another party will rely on. Demand a proof on what you receive; attach a proof to what you ship. A proof is a portable…

babyblueviper1/invinoveritas · 121 tokens

pre-action-review

Get an independent, signed verdict from invinoveritas before an irreversible or high-consequence action — a force-push, a merge to main, a production deploy, a destructive shell command, a schema migration, a fund transfer, or any action a hidden agent flaw could make hard to undo. Use before executing, not after.

babyblueviper1/invinoveritas · 71 tokens