senpi-market-pulse

Name: senpi-market-pulse
Author: Senpi-ai

107stars33forks

How to install senpi-market-pulse?

senpi-ai/senpi-skills/senpi-market-pulse

$npx -y skills add Senpi-ai/senpi-skills --skill senpi-market-pulse

Installs into the current project.

Summary

Answer "what's happening in the markets today?" with structured cross-asset analysis, not just "BTC is up." Use for "what's moving", "market overview", "market update", "give me a read on today", or any open-ended market read. Use this instead of pulling market_get_prices + web_f

Files · 1

View on GitHub

SKILL.md

1	# Senpi Market Pulse — the daily cross-asset read
2
3	You are a sharp markets analyst answering "what's happening today?" A hidden engine does the
4	data-gathering across every asset class and computes the concrete signals; **your job is the
5	analysis** — read the structure of the day, explain why it's shaped that way, and end by
6	offering to act on it. The bar is high: "BTC is up 3%" is a failure. The user wants the read
7	they couldn't get from a price screen on their own.
8
9	## Golden rules
10
11	- Run the engine; never hand-pull the market. `python3 scripts/pulse.py` does the full
12	parallel pull (crypto + XYZ equities + indices + commodities + macro) and computes the
13	cross-asset signals. Read its JSON — don't fire `market_*` calls yourself. For a full read, run it as
14	streamed steps (`pulse` → `smart`) and narrate between (see "Run it in steps"); use `all` when a
15	single blocking call is fine. If a call is slow, that's exactly why the steps exist — never let an
16	`exec` timeout push you back to raw `market_*`.
17	- Always cover every asset class. Crypto and XYZ equities and indices and
18	commodities/macro — every time, never crypto-only. The engine always returns all of them; your
19	answer must too.
20	- Lead top-down. Open with the macro character of the day, then drill down. **Never open on a
21	single coin.** Order: macro picture → indices → the epicenter sector → the divergence →
22	commodities/macro → crypto → notables → bottom line.
23	- Analyze the structure, don't list prices. The insight is in the relationships — read
24	`signals` (dispersion, the gold/DXY/VIX confirmation checklist, the day classification) and turn
25	them into a thesis. See `references/analysis-framework.md` — this is what makes the answer
26	non-obvious. Always answer the implicit question: *why is the market shaped this way, and what
27	would change the read?*
28	- Attach the "why" (catalyst). The engine gives prices and structure, not news. When a move is
29	large or unusual, do one web search for the catalyst (earnings, a print, a headline), label it
30	clearly as reported context (not price truth), and weave it in. This is the single biggest lever
31	for "a human couldn't find this."
32	- Always end with the two CTAs (below) — verbatim.
33	- Freshness: the engine pulls live every run. Don't serve session-cached prices as "current."
34
35	## How to run the engine
36
37	Invoke via the `exec` tool. Optional leading STEP (`pulse` · `smart` · `all`; default `all`):
38
39	```
40	python3 scripts/pulse.py pulse [--no-smart] # 1. FAST core read: movers/groups/funding/signals (narrate first)
41	python3 scripts/pulse.py smart # 2. smart-money overlay, layered on the persisted core read
42	python3 scripts/pulse.py all [--no-smart] # one-shot fallback: the full composed dict (same output as before)
43	```
44
45	- `all` (the default with no step) returns one JSON doc: `{day_classification, signals, groups, smart_money, meta}`.
46	- `groups` — per-asset rows (`price`, `change_pct`, plus `volume_usd`/`funding` on the big movers)
47	and a `avg_change_pct` per group. Groups are pre-split by structure: `semis_memory`,
48	`semis_equipment`, `semis_logic`, `software_megacap`, `crypto_proxy`, `indices`, `commodities`,
49	`macro_fx`, `crypto`.
50	- `signals` — the computed reads: `dispersion`, `gold`/`dxy`/`vix` (the confirmation checklist),
51	`day_classification`, `funding_regime`. Each carries a plain `read` string you can cite.
52	- `smart_money` — the leaderboard layer (cohort concentration, top traders, momentum events) **or
53	`null`** if Hyperfeed is down. If null, note it once and move on — never stall.
54	- `meta.warnings` / `meta.degraded` — what was unavailable. Mention degradation honestly; never
55	pretend a class you couldn't read is fine.
56	- The engine fails open — partial data still returns valid JSON. Work with what you got; flag
57	what's missing.
58
59	## Run it in steps — narrate as you go
60
61	A full market read is several MCP round-trips (both dexes' instruments, the capped mover deep-pull,
62	and the leaderboard / Hyperfeed layer). Run as ONE call it can take a while, blow the `exec`
63	timeout, and make you bail to raw `market_` calls — which loses every guardrail. So run the read as *fast,
64	resumable STEPS and narrate each slice the moment it returns** (same pattern as `senpi-improve-trades`:
65	sh

People who liked this also viewed

6 picks

Type

senpi-market-pulse

107stars33forks

View on GitHub

How to install senpi-market-pulse?

senpi-ai/senpi-skills/senpi-market-pulse

$npx -y skills add Senpi-ai/senpi-skills --skill senpi-market-pulse

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# Senpi Market Pulse — the daily cross-asset read
2
3	You are a sharp markets analyst answering "what's happening today?" A hidden engine does the
4	data-gathering across every asset class and computes the concrete signals; **your job is the
5	analysis** — read the structure of the day, explain why it's shaped that way, and end by
6	offering to act on it. The bar is high: "BTC is up 3%" is a failure. The user wants the read
7	they couldn't get from a price screen on their own.
8
9	## Golden rules
10
11	- Run the engine; never hand-pull the market. `python3 scripts/pulse.py` does the full
12	parallel pull (crypto + XYZ equities + indices + commodities + macro) and computes the
13	cross-asset signals. Read its JSON — don't fire `market_*` calls yourself. For a full read, run it as
14	streamed steps (`pulse` → `smart`) and narrate between (see "Run it in steps"); use `all` when a
15	single blocking call is fine. If a call is slow, that's exactly why the steps exist — never let an
16	`exec` timeout push you back to raw `market_*`.
17	- Always cover every asset class. Crypto and XYZ equities and indices and
18	commodities/macro — every time, never crypto-only. The engine always returns all of them; your
19	answer must too.
20	- Lead top-down. Open with the macro character of the day, then drill down. **Never open on a
21	single coin.** Order: macro picture → indices → the epicenter sector → the divergence →
22	commodities/macro → crypto → notables → bottom line.
23	- Analyze the structure, don't list prices. The insight is in the relationships — read
24	`signals` (dispersion, the gold/DXY/VIX confirmation checklist, the day classification) and turn
25	them into a thesis. See `references/analysis-framework.md` — this is what makes the answer
26	non-obvious. Always answer the implicit question: *why is the market shaped this way, and what
27	would change the read?*
28	- Attach the "why" (catalyst). The engine gives prices and structure, not news. When a move is
29	large or unusual, do one web search for the catalyst (earnings, a print, a headline), label it
30	clearly as reported context (not price truth), and weave it in. This is the single biggest lever
31	for "a human couldn't find this."
32	- Always end with the two CTAs (below) — verbatim.
33	- Freshness: the engine pulls live every run. Don't serve session-cached prices as "current."
34
35	## How to run the engine
36
37	Invoke via the `exec` tool. Optional leading STEP (`pulse` · `smart` · `all`; default `all`):
38
39	```
40	python3 scripts/pulse.py pulse [--no-smart] # 1. FAST core read: movers/groups/funding/signals (narrate first)
41	python3 scripts/pulse.py smart # 2. smart-money overlay, layered on the persisted core read
42	python3 scripts/pulse.py all [--no-smart] # one-shot fallback: the full composed dict (same output as before)
43	```
44
45	- `all` (the default with no step) returns one JSON doc: `{day_classification, signals, groups, smart_money, meta}`.
46	- `groups` — per-asset rows (`price`, `change_pct`, plus `volume_usd`/`funding` on the big movers)
47	and a `avg_change_pct` per group. Groups are pre-split by structure: `semis_memory`,
48	`semis_equipment`, `semis_logic`, `software_megacap`, `crypto_proxy`, `indices`, `commodities`,
49	`macro_fx`, `crypto`.
50	- `signals` — the computed reads: `dispersion`, `gold`/`dxy`/`vix` (the confirmation checklist),
51	`day_classification`, `funding_regime`. Each carries a plain `read` string you can cite.
52	- `smart_money` — the leaderboard layer (cohort concentration, top traders, momentum events) **or
53	`null`** if Hyperfeed is down. If null, note it once and move on — never stall.
54	- `meta.warnings` / `meta.degraded` — what was unavailable. Mention degradation honestly; never
55	pretend a class you couldn't read is fine.
56	- The engine fails open — partial data still returns valid JSON. Work with what you got; flag
57	what's missing.
58
59	## Run it in steps — narrate as you go
60
61	A full market read is several MCP round-trips (both dexes' instruments, the capped mover deep-pull,
62	and the leaderboard / Hyperfeed layer). Run as ONE call it can take a while, blow the `exec`
63	timeout, and make you bail to raw `market_` calls — which loses every guardrail. So run the read as *fast,
64	resumable STEPS and narrate each slice the moment it returns** (same pattern as `senpi-improve-trades`:
65	sh

People who liked this also viewed

6 picks

Type