一键导入
verify
Check which cloud image providers are configured and report readiness status.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Check which cloud image providers are configured and report readiness status.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Produce a labeled figure with PERFECT text — render a label-free image locally (or take an external image) then overlay leader lines and typeset labels programmatically from vision-derived anchor coordinates. Text is correct by construction; only pointer placement is reviewed.
Ask which image models fit a set of tasks — get evidence-based recommendations with per-image costs, which external services to pay for, and which local models to install (with disk/RAM/licence guidance). Reads the live model catalog and probes what is actually available on this machine.
Check mflux (MLX) availability, per-family runtime presence, cached weights, and report readiness status.
Top-level image orchestrator. Routes all slide image generation to the appropriate skill (jack-tar-ollama:image, jack-tar-ollama:icon, jack-tar-ollama:pattern, jack-tar-ollama:diagram, jack-tar-cloud:image, jack-tar-cloud:icon, render_chart). Produces ImageManifest and ChartManifest. Also reads strategy-map.json to determine per-slide rendering approach (full_render, backdrop_render, composed).
Meta-verify — discover all jack-tar engine plugins, call each verify, report aggregate pipeline capability and discipline-hook readiness.
Generate an image locally on Apple Silicon via the mflux CLI (MLX). Flag-compatible with jack-tar-ollama's /image — a $0 local tier, no API keys needed.
| name | verify |
| description | Check which cloud image providers are configured and report readiness status. |
| allowed-tools | Bash(python *) |
Check whether this plugin's prerequisites are met and report per-provider readiness.
python3 -c "
missing = []
try:
import requests
except ImportError:
missing.append('requests')
try:
import openai
except ImportError:
missing.append('openai')
try:
from google import genai
except ImportError:
missing.append('google-genai')
try:
import fal_client
except ImportError:
missing.append('fal-client')
if missing:
print('MISSING:' + ','.join(missing))
else:
print('ALL_PRESENT')
"
If any packages are missing, report NOT_AVAILABLE with install instructions (pip install <package>).
python3 -c "
import os
providers = {
'openai': bool(os.environ.get('OPENAI_API_KEY')),
'google': bool(os.environ.get('GOOGLE_CLOUD_PROJECT') or os.environ.get('GOOGLE_API_KEY')),
'fal': bool(os.environ.get('FAL_KEY')),
}
# Recraft uses OPENAI_API_KEY
providers['recraft'] = providers['openai']
import json
print(json.dumps(providers))
"
Count ready providers and report one of:
STATUS: FULLY_AVAILABLE — all dependencies present and at least 2 providers configuredSTATUS: PARTIALLY_AVAILABLE — dependencies present but only 1-3 providers configuredSTATUS: NOT_AVAILABLE — missing Python dependencies or zero providers configuredExample output:
PLUGIN: jack-tar-cloud
VERSION: 1.0.0
DEPENDENCIES:
Python: READY (3.12.x)
requests: READY
openai: READY
google-genai: READY
fal-client: READY
PROVIDERS:
openai: READY (OPENAI_API_KEY set)
google-nanobanana: READY (Flash + Pro via GOOGLE_API_KEY)
google-imagen: READY (Fast + Standard via GOOGLE_API_KEY)
fal: READY (FAL_KEY set)
recraft: READY (uses OPENAI_API_KEY)
GOOGLE TIERS:
Nanobanana Flash: gemini-3.1-flash-image $0.067/image (best text rendering)
Nanobanana Pro: gemini-3-pro-image $0.134/image (premium quality)
Imagen Fast: imagen-4.0-fast-generate-001 $0.020/image (budget bulk)
Imagen Standard: imagen-4.0-generate-001 $0.040/image (standard quality)
CAPABILITIES:
image: READY (3/4 providers available)
icon: READY (recraft available)
STATUS: PARTIALLY_AVAILABLE
REASON: All providers configured
Report which model catalog is in effect and how old the local cache is:
python3 - << 'PYEOF'
import sys
sys.path.insert(0, "PLUGIN_ROOT") # replace with the plugin root path
from src.model_catalog_refresh import staleness_report
report = staleness_report()
print("MODEL CATALOG:")
print(f" version: {report['version']} (data updated {report['updated']})")
print(f" source: {report['source']}")
if "cache_age_days" in report:
print(f" cache age: {report['cache_age_days']} days")
if report["cache_age_days"] > 30:
print(" hint: cache is over a month old — run /jack-tar-cloud:refresh-models")
else:
print(" hint: running on the shipped baseline — /jack-tar-cloud:refresh-models fetches the latest")
PYEOF
This never blocks verification — model ids and prices move faster than plugin releases (EPIC #125), so the hint just tells the operator when a refresh is worth running.
When the operator asks to "check the models", "find the latest models", or after any 404/NOT_FOUND from a provider, run the live discovery probe. It asks each probeable provider (Google, OpenAI, Ollama) what it serves RIGHT NOW and classifies every catalog entry:
python3 - << 'PYEOF'
import json, sys
sys.path.insert(0, "PLUGIN_ROOT") # replace with the plugin root path
from src.model_probe import probe_report
report = probe_report()
print(json.dumps(report, indent=2))
PYEOF
Present the results as:
/jack-tar-cloud:refresh-models (a corrected catalog may already be
published) or a local-config override.model_catalog override in local-config.json
(explicit opt-in) or propose for the canonical catalog.