senpi-trading-runtime

Name: senpi-trading-runtime
Author: Senpi-ai

107stars33forks

How to install senpi-trading-runtime?

senpi-ai/senpi-skills/senpi-trading-runtime

$npx -y skills add Senpi-ai/senpi-skills --skill senpi-trading-runtime

Installs into the current project.

Summary

How a Senpi trading strategy interacts with the runtime engine (@senpi-ai/runtime) on Hyperliquid: a strategy runs from a runtime.yaml pointing at a Python module that exports scan(inputs, ctx), which the runtime supervises and calls each interval, then owns execution, risk guard

Files · 1

View on GitHub

SKILL.md

1	# Senpi Trading Runtime — the runtime contract
2
3	This skill is infrastructure: the canonical knowledge of how the Senpi runtime
4	(`@senpi-ai/runtime`) behaves and how a strategy interacts with it. The lifecycle skills —
5	author (build), ops (install/monitor), discover (recommend) — reference this one for the contract.
6
7	## The runtime model
8
9	A strategy runs from a `runtime.yaml` that points at an in-repo Python module. The runtime **spawns and
10	supervises that module and calls a frozen `scan(inputs, ctx)`** every `interval_seconds`. The
11	division of labor is fixed:
12
13	- Your code produces signals — nothing else. `scan(inputs, ctx)` reads market and account data
14	and returns a `list[dict]` of candidate signals. It does not open, close, size, schedule, or
15	execute anything.
16	- The runtime owns everything downstream: scheduling (`interval_seconds`), spawning +
17	supervising + restarting the scanner, validating (`signal_data_schema`) + de-duplicating the
18	signals you return, sizing & order execution (`FEE_OPTIMIZED_LIMIT`), slot accounting,
19	`risk.guard_rails`, the two-phase DSL trailing-stop exits, and crash-safe position reconcile
20	on restart.
21
22	## How your code talks to the runtime
23
24	The interaction surface is small and one-directional — you read, you return signals, the runtime acts.
25
26	- `runtime.yaml` declares the scanner(s), the action gate, the exit engine, and the risk
27	guard-rails, and passes author tunables down via `inputs:`. → `references/runtime-yaml.md`
28	- `scan(inputs, ctx)` is the single entry point. `inputs` is the runtime's `inputs:` map; `ctx`
29	gives you:
30	- `ctx.senpi_mcp.call_tool(name, args)` — the Senpi MCP client, read-only (market,
31	account, leaderboard, discovery, `strategy_get*`, …). It is the only way to fetch data.
32	- `ctx.state` — transactional, runtime-persisted history (`last()` / `append()` / `len`) for
33	dedup, rotation, and first-seen ledgers; advances only on a clean tick.
34	- `ctx.wallet` — the strategy's wallet address.
35	- → `references/scan-contract.md`
36	- The return value is a `list[dict]`, one per candidate signal (`asset`, `direction`,
37	`marginUsd`, `leverage`, `data{}`). The runtime validates each `data{}` against the runtime.yaml's
38	`signal_data_schema`, then sizes, executes, and manages exits.
39
40	Keep the thesis logic in a sibling pure `scoring.py` (no I/O, no MCP) so it is unit-testable;
41	`scan.py` does the reads + state, `scoring.py` does the math.
42
43	## Runtime commands (essentials)
44
45	The plugin registers a `senpi` command group on the gateway. Deploying a runtime and checking it:
46
47	```bash
48	openclaw plugins install @senpi-ai/runtime
49	openclaw senpi runtime create -p runtime.yaml # hot-loads; the runtime supervises the scanner
50	openclaw senpi runtime list # id, source, status (running/stopped)
51	```
52
53	Beyond `runtime create/list/delete`, the CLI exposes the runtime's live state — `senpi dsl
54	positions\|inspect\|closes` (the exit engine), `senpi action list\|inspect\|history\|decisions` (the
55	decision layer), `senpi risk` (am I allowed to trade, and why not), `senpi audit` (backend trade
56	trail with AI reasoning), `senpi scanner` (per-scanner health, liveness, and barren detection),
57	`senpi events`/`senpi explain <asset>` (the local domain-event log — the trade narrative, and one
58	asset's stitched lifecycle), `senpi status`/`senpi state` (health), and `senpi guide …` (in-shell
59	reference). Full surface with every option → `references/runtime-cli.md`.
60
61	To confirm open positions are actually stop-loss protected (a position with no DSL shows up as an
62	absence in `dsl positions`, so it's easy to miss) → the verdict procedure in
63	`references/dsl-protection-check.md`.
64
65	## The reference set
66
67	\| Read this \| For \|
68	\|---\|---\|
69	\| `references/runtime-concepts.md` \| How the runtime behaves end to end: the runtime pipeline, `position_tracker`, and the two-phase DSL exit engine \|
70	\| `references/runtime-yaml.md` \| The `runtime.yaml` schema — every section, the `external_scanner` fields, the risk guard-rails \|
71	\| `references/scan-contract

People who liked this also viewed

6 picks

Type

senpi-trading-runtime

107stars33forks

View on GitHub

How to install senpi-trading-runtime?

senpi-ai/senpi-skills/senpi-trading-runtime

$npx -y skills add Senpi-ai/senpi-skills --skill senpi-trading-runtime

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Senpi Trading Runtime — the runtime contract
2
3	This skill is infrastructure: the canonical knowledge of how the Senpi runtime
4	(`@senpi-ai/runtime`) behaves and how a strategy interacts with it. The lifecycle skills —
5	author (build), ops (install/monitor), discover (recommend) — reference this one for the contract.
6
7	## The runtime model
8
9	A strategy runs from a `runtime.yaml` that points at an in-repo Python module. The runtime **spawns and
10	supervises that module and calls a frozen `scan(inputs, ctx)`** every `interval_seconds`. The
11	division of labor is fixed:
12
13	- Your code produces signals — nothing else. `scan(inputs, ctx)` reads market and account data
14	and returns a `list[dict]` of candidate signals. It does not open, close, size, schedule, or
15	execute anything.
16	- The runtime owns everything downstream: scheduling (`interval_seconds`), spawning +
17	supervising + restarting the scanner, validating (`signal_data_schema`) + de-duplicating the
18	signals you return, sizing & order execution (`FEE_OPTIMIZED_LIMIT`), slot accounting,
19	`risk.guard_rails`, the two-phase DSL trailing-stop exits, and crash-safe position reconcile
20	on restart.
21
22	## How your code talks to the runtime
23
24	The interaction surface is small and one-directional — you read, you return signals, the runtime acts.
25
26	- `runtime.yaml` declares the scanner(s), the action gate, the exit engine, and the risk
27	guard-rails, and passes author tunables down via `inputs:`. → `references/runtime-yaml.md`
28	- `scan(inputs, ctx)` is the single entry point. `inputs` is the runtime's `inputs:` map; `ctx`
29	gives you:
30	- `ctx.senpi_mcp.call_tool(name, args)` — the Senpi MCP client, read-only (market,
31	account, leaderboard, discovery, `strategy_get*`, …). It is the only way to fetch data.
32	- `ctx.state` — transactional, runtime-persisted history (`last()` / `append()` / `len`) for
33	dedup, rotation, and first-seen ledgers; advances only on a clean tick.
34	- `ctx.wallet` — the strategy's wallet address.
35	- → `references/scan-contract.md`
36	- The return value is a `list[dict]`, one per candidate signal (`asset`, `direction`,
37	`marginUsd`, `leverage`, `data{}`). The runtime validates each `data{}` against the runtime.yaml's
38	`signal_data_schema`, then sizes, executes, and manages exits.
39
40	Keep the thesis logic in a sibling pure `scoring.py` (no I/O, no MCP) so it is unit-testable;
41	`scan.py` does the reads + state, `scoring.py` does the math.
42
43	## Runtime commands (essentials)
44
45	The plugin registers a `senpi` command group on the gateway. Deploying a runtime and checking it:
46
47	```bash
48	openclaw plugins install @senpi-ai/runtime
49	openclaw senpi runtime create -p runtime.yaml # hot-loads; the runtime supervises the scanner
50	openclaw senpi runtime list # id, source, status (running/stopped)
51	```
52
53	Beyond `runtime create/list/delete`, the CLI exposes the runtime's live state — `senpi dsl
54	positions\|inspect\|closes` (the exit engine), `senpi action list\|inspect\|history\|decisions` (the
55	decision layer), `senpi risk` (am I allowed to trade, and why not), `senpi audit` (backend trade
56	trail with AI reasoning), `senpi scanner` (per-scanner health, liveness, and barren detection),
57	`senpi events`/`senpi explain <asset>` (the local domain-event log — the trade narrative, and one
58	asset's stitched lifecycle), `senpi status`/`senpi state` (health), and `senpi guide …` (in-shell
59	reference). Full surface with every option → `references/runtime-cli.md`.
60
61	To confirm open positions are actually stop-loss protected (a position with no DSL shows up as an
62	absence in `dsl positions`, so it's easy to miss) → the verdict procedure in
63	`references/dsl-protection-check.md`.
64
65	## The reference set
66
67	\| Read this \| For \|
68	\|---\|---\|
69	\| `references/runtime-concepts.md` \| How the runtime behaves end to end: the runtime pipeline, `position_tracker`, and the two-phase DSL exit engine \|
70	\| `references/runtime-yaml.md` \| The `runtime.yaml` schema — every section, the `external_scanner` fields, the risk guard-rails \|
71	\| `references/scan-contract

People who liked this also viewed

6 picks

Type