senpi-strategy-ops
starsforks
How to install senpi-strategy-ops?
senpi-ai/senpi-skills/senpi-strategy-ops
$npx -y skills add Senpi-ai/senpi-skills --skill senpi-strategy-opsDeploy / monitor / close a NAMED Senpi trading strategy. Use when the user names a strategy to run — "install spider", "deploy polar", "set up kodiak", "run the spider strategy", "is my strategy live?", "what am I running", "list my strategies" (→ status.py), "are my positions pr
SKILL.md
| 1 | # Senpi Strategy Ops — deploy, monitor, close |
| 2 | |
| 3 | A strategy is a **package**: `strategy.yaml` (the deploy manifest) + one `runtime.yaml` per instance + |
| 4 | `<instance>/scanners/`. The runtime **spawns and supervises** each `scanners/scan.py`, calling |
| 5 | `scan(inputs, ctx)` every `interval_seconds` and owning everything downstream — signal validation, |
| 6 | sizing/execution, the two-phase DSL exit, risk guard-rails. **There is no separate scanner daemon, no |
| 7 | `push_signal`.** Ops owns the deployed lifecycle. Deploy is **three short, resumable steps** (each fits a |
| 8 | tool call — wallet funding and the first scan tick are slow, so they must not block one long call): |
| 9 | |
| 10 | ``` |
| 11 | python3 senpi-strategy-ops/scripts/deploy.py validate <id> # 0. preflight — deploy-ready? (no side effects) |
| 12 | python3 senpi-strategy-ops/scripts/deploy.py create <id> --budget <usd> # 1. create wallets & fund them |
| 13 | python3 senpi-strategy-ops/scripts/deploy.py runtime <id> # 2. register the runtime(s) |
| 14 | python3 senpi-strategy-ops/scripts/deploy.py verify <id> # 3. GATE — confirm LIVE (runtime+scanner+DSL+budget) |
| 15 | python3 senpi-strategy-ops/scripts/status.py # what am I running? (+ health) |
| 16 | python3 senpi-strategy-ops/scripts/close.py <id> # teardown one strategy |
| 17 | python3 senpi-strategy-ops/scripts/close.py --all # teardown EVERY open strategy |
| 18 | ``` |
| 19 | **Always tear down through `close.py`** (one `<id>` or `--all`) — it deletes the runtime *and* closes the |
| 20 | strategy. A raw `strategy_close` MCP call closes the strategy but **leaves the runtime registered**, which |
| 21 | collides on the next deploy. "close all strategies / return funds to main" → `close.py --all`. |
| 22 | Pass the **strategy `id`** for a CATALOG strategy (what `senpi-strategy-discover` hands over, e.g. |
| 23 | `spider`) — it's fetched from the remote if not on disk. For a **locally-authored package, pass its |
| 24 | DIRECTORY path** (absolute is safest, e.g. `/data/workspace/strategies/<id>`): a bare id only resolves |
| 25 | locally relative to your CWD, so from anywhere else it becomes a remote catalog fetch — never what you |
| 26 | want for a package you just wrote. An on-disk package is authoritative; an invalid one surfaces its |
| 27 | real errors and is never silently replaced by a remote fetch. The scripts call MCP directly |
| 28 | (`scripts/mcp_client.py`, reads |
| 29 | `SENPI_AUTH_TOKEN`) + drive `openclaw senpi runtime …`. Mechanics + state machine: |
| 30 | [`references/lifecycle.md`](references/lifecycle.md). Manifest: [`references/strategy-yaml-schema.md`](references/strategy-yaml-schema.md). |
| 31 | |
| 32 | ## Deploy — three steps: create → runtime → verify (NOT live until `verify` passes) |
| 33 | |
| 34 | > **The loop is a gate, not a suggestion.** A strategy is LIVE only when `verify` returns `live` — every |
| 35 | > instance **runtime-running + scanner-active + DSL-wired + funded to what was requested**. If any step is |
| 36 | > incomplete, the strategy is **not live**; say so and fix the flagged component. Never report "live" off |
| 37 | > `registered` alone. |
| 38 | |
| 39 | **Step 0 — resolve which strategy.** The user's word ("spider") is a strategy **`id`**. To confirm it |
| 40 | exists, check the registry; no match → hand to **senpi-strategy-discover**: |
| 41 | ``` |
| 42 | curl -s https://raw.githubusercontent.com/Senpi-ai/senpi-skills/refs/heads/main/strategies/catalog.json |
| 43 | ``` |
| 44 | |
| 45 | **Step 0.5 — preflight (recommended).** `deploy.py validate <id>` reports every structural + render |
| 46 | issue in **one pass**, with **no side effects**, before you fund anything. The deployer **accepts the |
| 47 | flat single-instance layout** agents naturally scaffold — one `runtime.yaml` + `scanners/` at the |
| 48 | package root — by synthesizing the canonical `main` instance, so there's **no need to restructure into |
| 49 | `main/`**. Any remaining fix is named prescriptivel |
People who liked this also viewed
6 picksType
- azure-costAzure cost management: query costs, forecast spending, optimize to reduce waste. WHEN: \"Azure costs\", \"Azure bill\", \"cost breakdown\", \"how much am I spending\", \"forecast spending\", \"optimizSkillsJul 2026
- pricingWhen the user wants help with pricing decisions, packaging, or monetization strategy. Also use when the user mentions 'pricing,' 'pricing tiers,' 'freemium,' 'free trial,' 'packaging,' 'price increaseSkillsJul 2026
- stripe-integrationMulti-harness agentic plugin marketplace for Claude Code, Codex CLI, Cursor, OpenCode, GitHub Copilot, and Gemini CLISkillsJul 2026
- startup-financial-modelingMulti-harness agentic plugin marketplace for Claude Code, Codex CLI, Cursor, OpenCode, GitHub Copilot, and Gemini CLISkillsJul 2026
- tushare-finance专为 Claude Code 设计的 Tushare Pro 金融数据技能包，支持 220+ 个接口，用自然语言获取 A 股行情、财务数据、宏观经济等金融数据。SkillsJul 2026
- azure-pricingFetches real-time Azure retail pricing using the Azure Retail Prices API (prices.azure.com) and estimates Copilot Studio agent credit consumption. Use when the user asks about the cost of any Azure seSkillsJul 2026
