refactor: split precheck state snapshot and analysis

This commit is contained in:
2026-04-15 17:17:00 +08:00
parent 965b53b598
commit 893f0fb077
6 changed files with 287 additions and 57 deletions

214
README.md
View File

@@ -1,13 +1,69 @@
# coinhunter-cli
CoinHunter CLI is the executable tooling layer for CoinHunter.
<p align="center">
<strong>The executable CLI layer for CoinHunter.</strong><br/>
Runtime-safe trading operations, precheck orchestration, review tooling, and market probes.
</p>
• Code lives in this repository.
• User runtime data lives in ~/.coinhunter/ by default.
• Hermes skills can call this CLI instead of embedding large script collections.
• Runtime locations can be overridden with COINHUNTER_HOME, HERMES_HOME, COINHUNTER_ENV_FILE, and HERMES_BIN.
<p align="center">
<img alt="Python" src="https://img.shields.io/badge/python-3.10%2B-blue" />
<img alt="Status" src="https://img.shields.io/badge/status-active%20development-orange" />
<img alt="Architecture" src="https://img.shields.io/badge/architecture-runtime%20%2B%20commands%20%2B%20services-6f42c1" />
</p>
## Install
## Why this repo exists
CoinHunter is evolving from a loose bundle of automation scripts into a proper installable command-line tool.
This repository is the tooling layer:
- Code and executable behavior live here.
- User runtime state lives in `~/.coinhunter/` by default.
- Hermes skills can call this CLI instead of embedding large script collections.
- Runtime paths can be overridden with `COINHUNTER_HOME`, `HERMES_HOME`, `COINHUNTER_ENV_FILE`, and `HERMES_BIN`.
In short:
- `coinhunter-cli` = tool
- CoinHunter skill = strategy / workflow / prompting layer
- `~/.coinhunter` = user data, logs, state, reviews
## Current architecture
```text
coinhunter-cli/
├── src/coinhunter/
│ ├── cli.py # top-level command router
│ ├── runtime.py # runtime paths + env loading
│ ├── doctor.py # diagnostics
│ ├── paths.py # runtime path inspection
│ ├── commands/ # thin CLI adapters
│ ├── services/ # orchestration / application services
│ └── *.py # compatibility modules + legacy logic under extraction
└── README.md
```
The repo is actively being refactored toward a cleaner split:
- `commands/` → argument / CLI adapters
- `services/` → orchestration and application workflows
- `runtime/` → paths, env, files, locks, config
- future `domain/` → trading and precheck core logic
## Implemented command/service splits
The first extraction pass is already live:
- `smart-executor``commands.smart_executor` + `services.smart_executor_service`
- `precheck``commands.precheck` + `services.precheck_service`
- `precheck` internals now also have dedicated service modules for:
- `services.precheck_state`
- `services.precheck_snapshot`
- `services.precheck_analysis`
This keeps behavior stable while giving the codebase a cleaner landing zone for deeper refactors.
## Installation
Editable install:
@@ -15,38 +71,152 @@ Editable install:
pip install -e .
```
## Core commands
Run directly after install:
```bash
coinhunter --help
coinhunter --version
coinhunter doctor
coinhunter paths
```
## Quickstart
Initialize user state:
```bash
coinhunter init
```
Inspect runtime wiring:
```bash
coinhunter paths
coinhunter doctor
```
Validate exchange credentials:
```bash
coinhunter check-api
coinhunter smart-executor balances
```
Run precheck / gate plumbing:
```bash
coinhunter precheck
coinhunter precheck --mark-run-requested "external-gate queued cron run"
coinhunter precheck --ack "analysis finished"
```
Inspect balances or execute trading actions:
```bash
coinhunter smart-executor balances
coinhunter smart-executor status
coinhunter smart-executor hold
coinhunter smart-executor buy ENJUSDT 50
coinhunter smart-executor sell-all ENJUSDT
```
Generate review data:
```bash
coinhunter review-context 12
coinhunter review-engine 12
```
Probe external market data:
```bash
coinhunter market-probe bybit-ticker BTCUSDT
coinhunter market-probe bybit-klines BTCUSDT 60 20
```
## Runtime model
Default layout:
• ~/.coinhunter/ stores config, positions, logs, reviews, and state.
~/.hermes/.env stores exchange credentials unless COINHUNTER_ENV_FILE overrides it.
• hermes is discovered from PATH first, then ~/.local/bin/hermes, unless HERMES_BIN overrides it.
```text
~/.coinhunter/
├── accounts.json
├── config.json
├── executions.json
├── notes.json
├── positions.json
├── watchlist.json
├── logs/
├── reviews/
└── state/
```
## Next refactor direction
Credential loading:
This repository now has a dedicated runtime layer and CLI diagnostics. The next major cleanup is to split command adapters from trading services so the internal architecture becomes:
- Binance credentials are read from `~/.hermes/.env` by default.
- `COINHUNTER_ENV_FILE` can point to a different env file.
- `hermes` is resolved from `PATH` first, then `~/.local/bin/hermes`, unless `HERMES_BIN` overrides it.
commands/
• services/
• runtime/
• domain logic
## Useful commands
The first split is already in place for:
### Diagnostics
• smart-executor -> commands.smart_executor + services.smart_executor_service
• precheck -> commands.precheck + services.precheck_service
```bash
coinhunter doctor
coinhunter paths
coinhunter check-api
```
### Trading and execution
```bash
coinhunter smart-executor balances
coinhunter smart-executor status
coinhunter smart-executor hold
coinhunter smart-executor rebalance FROMUSDT TOUSDT
```
### Precheck and orchestration
```bash
coinhunter precheck
coinhunter external-gate
coinhunter rotate-external-gate-log
```
### Review and market research
```bash
coinhunter review-context 12
coinhunter review-engine 12
coinhunter market-probe bybit-ticker BTCUSDT
```
## Development notes
This project is intentionally moving in small, safe refactor steps:
1. Separate runtime concerns from hardcoded paths.
2. Move command dispatch into thin adapters.
3. Introduce orchestration services.
4. Extract reusable domain logic from large compatibility modules.
5. Keep cron / Hermes integration stable during migration.
That means some compatibility modules still exist, but the direction is deliberate.
## Near-term roadmap
- Extract more logic from `smart_executor.py` into dedicated execution / portfolio services.
- Continue shrinking `precheck.py` by moving snapshot and analysis internals into reusable modules.
- Add `domain/` models for positions, signals, and trigger analysis.
- Add tests for runtime paths, precheck service behavior, and CLI stability.
- Evolve toward a more polished installable CLI workflow.
## Philosophy
CoinHunter should become:
- more professional
- more maintainable
- safer to operate
- easier for humans and agents to call
- less dependent on prompt-only correctness
This repo is where that evolution happens.

View File

@@ -904,23 +904,15 @@ def update_state_after_observation(state: dict, snapshot: dict, analysis: dict):
def mark_run_requested(note: str = ""):
state = load_state()
state["run_requested_at"] = utc_iso()
state["run_request_note"] = note
save_state(state)
print(json.dumps({"ok": True, "run_requested_at": state["run_requested_at"], "note": note}, ensure_ascii=False))
from .services.precheck_state import mark_run_requested as service_mark_run_requested
return service_mark_run_requested(note)
def ack_analysis(note: str = ""):
state = load_state()
state["last_deep_analysis_at"] = utc_iso()
state["pending_trigger"] = False
state["pending_reasons"] = []
state["last_ack_note"] = note
state.pop("run_requested_at", None)
state.pop("run_request_note", None)
save_state(state)
print(json.dumps({"ok": True, "acked_at": state["last_deep_analysis_at"], "note": note}, ensure_ascii=False))
from .services.precheck_state import ack_analysis as service_ack_analysis
return service_ack_analysis(note)
def main():

View File

@@ -0,0 +1,25 @@
"""Analysis helpers for precheck."""
from __future__ import annotations
from .. import precheck as precheck_module
def analyze_trigger(snapshot: dict, state: dict) -> dict:
return precheck_module.analyze_trigger(snapshot, state)
def build_failure_payload(exc: Exception) -> dict:
return {
"generated_at": precheck_module.utc_iso(),
"status": "deep_analysis_required",
"should_analyze": True,
"pending_trigger": True,
"cooldown_active": False,
"reasons": ["precheck-error"],
"hard_reasons": ["precheck-error"],
"soft_reasons": [],
"soft_score": 0,
"details": [str(exc)],
"compact_summary": f"预检查失败,转入深度分析兜底: {exc}",
}

View File

@@ -5,39 +5,26 @@ from __future__ import annotations
import json
import sys
from .. import precheck as precheck_module
from . import precheck_analysis, precheck_snapshot, precheck_state
def run(argv: list[str] | None = None) -> int:
argv = list(sys.argv[1:] if argv is None else argv)
if argv and argv[0] == "--ack":
precheck_module.ack_analysis(" ".join(argv[1:]).strip())
precheck_state.ack_analysis(" ".join(argv[1:]).strip())
return 0
if argv and argv[0] == "--mark-run-requested":
precheck_module.mark_run_requested(" ".join(argv[1:]).strip())
precheck_state.mark_run_requested(" ".join(argv[1:]).strip())
return 0
try:
state = precheck_module.sanitize_state_for_stale_triggers(precheck_module.load_state())
snapshot = precheck_module.build_snapshot()
analysis = precheck_module.analyze_trigger(snapshot, state)
precheck_module.save_state(precheck_module.update_state_after_observation(state, snapshot, analysis))
state = precheck_state.sanitize_state_for_stale_triggers(precheck_state.load_state())
snapshot = precheck_snapshot.build_snapshot()
analysis = precheck_analysis.analyze_trigger(snapshot, state)
precheck_state.save_state(precheck_state.update_state_after_observation(state, snapshot, analysis))
print(json.dumps(analysis, ensure_ascii=False, indent=2))
return 0
except Exception as exc:
failure = {
"generated_at": precheck_module.utc_iso(),
"status": "deep_analysis_required",
"should_analyze": True,
"pending_trigger": True,
"cooldown_active": False,
"reasons": ["precheck-error"],
"hard_reasons": ["precheck-error"],
"soft_reasons": [],
"soft_score": 0,
"details": [str(exc)],
"compact_summary": f"预检查失败,转入深度分析兜底: {exc}",
}
print(json.dumps(failure, ensure_ascii=False, indent=2))
print(json.dumps(precheck_analysis.build_failure_payload(exc), ensure_ascii=False, indent=2))
return 0

View File

@@ -0,0 +1,9 @@
"""Snapshot construction helpers for precheck."""
from __future__ import annotations
from .. import precheck as precheck_module
def build_snapshot() -> dict:
return precheck_module.build_snapshot()

View File

@@ -0,0 +1,47 @@
"""State helpers for precheck orchestration."""
from __future__ import annotations
import json
from .. import precheck as precheck_module
def load_state() -> dict:
return precheck_module.load_state()
def save_state(state: dict) -> None:
precheck_module.save_state(state)
def sanitize_state_for_stale_triggers(state: dict) -> dict:
return precheck_module.sanitize_state_for_stale_triggers(state)
def update_state_after_observation(state: dict, snapshot: dict, analysis: dict) -> dict:
return precheck_module.update_state_after_observation(state, snapshot, analysis)
def mark_run_requested(note: str = "") -> dict:
state = load_state()
state["run_requested_at"] = precheck_module.utc_iso()
state["run_request_note"] = note
save_state(state)
payload = {"ok": True, "run_requested_at": state["run_requested_at"], "note": note}
print(json.dumps(payload, ensure_ascii=False))
return payload
def ack_analysis(note: str = "") -> dict:
state = load_state()
state["last_deep_analysis_at"] = precheck_module.utc_iso()
state["pending_trigger"] = False
state["pending_reasons"] = []
state["last_ack_note"] = note
state.pop("run_requested_at", None)
state.pop("run_request_note", None)
save_state(state)
payload = {"ok": True, "acked_at": state["last_deep_analysis_at"], "note": note}
print(json.dumps(payload, ensure_ascii=False))
return payload