ワンクリックで
187web-manifest
Use when compiling the 187SKILLS manifest, selecting a prompt, or routing a Layer-4 viral/elegant prompt.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Use when compiling the 187SKILLS manifest, selecting a prompt, or routing a Layer-4 viral/elegant prompt.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Versioning, changelogs, releases, migration notes, compatibility, deprecations, and adapter sync.
Use when routing across the full 187SKILLS suite or the NATASHA multi-agent integration stack (THREAD, COMPRESS, TENSION, SPARK, CORD, SCOUT, LAB, FUSE).
Use for EVM-first smart-contract assurance, DeFi economic risk analysis, severity/confidence separation, and responsible disclosure (no live keys or exploits).
Use for quantum algorithm selection, circuit design, optimization, resource estimation, and claim-disciplined audits (no unsupported advantage claims).
Use when generating a new repository, deploying to GitHub, creating an installer site, or routing the 187SKILLS suite.
Use when adding delight, micro-interactions, onboarding, community, or retention to a 187WEB project.
| name | 187web-manifest |
| description | Use when compiling the 187SKILLS manifest, selecting a prompt, or routing a Layer-4 viral/elegant prompt. |
| model_adapter | chatgpt |
ChatGPT adapter: Use the markdown below as the custom GPT / system instructions. Source:
../../.claude/skills/187web-manifest/SKILL.md.
Suite: Charlotte v2 stack. Parent:
187web-ecosystem · Compiler:
scripts/187web-compiler.sh · Registry: references/MANIFEST.xml
The manifest is the Library of Skills — 27 high-density XML-compiled prompts ready for the pre-prompt-compiler to inject into agentic sessions. One manifest, zero drift across long-run coding sessions.
Load this skill when you need to select, compile, or inject a specific prompt; initialize a long-run session; or route Layer 4 viral prompts (UI audit, A11y, XSS, state machines).
In addition to the 27 manifest prompts, the registry now routes the 16 first-class
public skills and 2 suite-wide subskills defined in docs/SKILL-CONTRACT.md. The
compiler treats each skill as a routable directive with its own frontmatter,
references, and templates.
Canonical skill roster: 187REPO 187CRAFT 187VIBE 187LAUNCH 187FREE 187RESEARCH 187SEO 187REVENUE 187DOCS 187WRITE 187LEARN 187TEST 187ACCESS+ 187INCLUDE 187VERSION 187PUBLISH.
PLAN.md, inject compiled JSON.ui-aesthetic-auditor, task-identification-extraction).| Layer | Name | Count | Routes to |
|---|---|---|---|
| 1 | Native OS Productivity | 5 | widow-weaver |
| 2 | Corporate & Workflow | 4 | widow-weaver (+ {{SelectedText}}) |
| 3 | Swarm-Mind Engineering | 6 | swarm-mind personas |
| 4 | Viral & Elegant Web/Agent | 12 | 187web-manifest + 187WEBDEV suite |
Canonical XML: references/MANIFEST.xml
Build plan: references/BUILD-PLAN.md
Phase roadmap: references/PHASE-ROADMAP.md (187webdev-templates mapped)
# Install registry (once)
mkdir -p ~/.187web/prompts
cp references/MANIFEST.xml ~/.187web/prompts/
# List all prompt IDs
./scripts/187web-compiler.sh --list
# Auto-detect power mode + folder → compile JSON
./scripts/187web-compiler.sh
# Explicit prompt
./scripts/187web-compiler.sh --prompt a11y-linting-agent
# Low-power override
E187WEB_POWER_MODE=low ./scripts/187web-compiler.sh
.\scripts\187web-compiler.ps1 -List
.\scripts\187web-compiler.ps1 -Prompt ui-aesthetic-auditor
.\scripts\187web-compiler.ps1 -Write -Emit # persist + push to OmniQube relay
$env:E187WEB_POWER_MODE = 'low'; .\scripts\187web-compiler.ps1
# Full session init (PLAN.md + relay + compile)
.\scripts\session-init.ps1
# 1. Install compiler hook (once)
.\scripts\install-compiler-hook.ps1
# 2. Start telemetry relay
node .\scripts\telemetry-relay.mjs
# 3. Open OmniQube Render Matrix
# http://localhost:3000/omniqube.html (with next dev)
{
"ecosystem": "187WEB",
"power_mode": "high",
"prompt_id": "ui-aesthetic-auditor",
"skill": "187webdesign",
"persona": "ui-spinner",
"directive": "...",
"neuro_toxin": { "toxicity": 0.3, "lethality": "medium" }
}
| ID | Alias | Use |
|---|---|---|
document-summarization | TLDR_Toxin | Dense docs → bullets |
linguistic-translation | Polyglot_Thread | Translate preserving structure |
tone-adjustment-polishing | Tone_Polish | Assertive professional rewrite |
pdf-dialogue | PDF_QA | Q&A against active document |
generative-drafting | Draft_Venom | SME draft from brief |
| ID | Alias | Vars |
|---|---|---|
email-follow-up | Email_FollowUp | {{SelectedText}} |
task-identification-extraction | Task_Extractor | {{SelectedText}} |
stylistic-unification | Voice_Unify | {{SelectedText}} |
annual-report-synthesizer | Report_Toxin | {{SelectedText}} |
| ID | Persona | Folder |
|---|---|---|
ml-systems-architect | alpha-architect | /MachineLearning |
llm-deployment-architect | rag-weaver | /RAG |
multimodal-agent-designer | ui-spinner | /Design |
s-agent-spatial-architect | spatial-architect | /Spatial |
edge-ai-deployment-specialist | edge-venom | /Hardware |
ai-ethics-reviewer | ethics-reviewer | /Ethics |
| ID | Pairs with |
|---|---|
ui-aesthetic-auditor | 187webdesign |
user-journey-mapper | agent flow design |
component-library-weaver | 187webdev-design-system |
seo-semantic-optimizer | DOM/meta audit |
a11y-linting-agent | 187webdev-qa |
persona-simulation-testing | {{Persona}} var |
copy-conversion-linter | 187webdesign |
responsive-breakpoint-debugger | CSS grid/flex |
npu-performance-linter | edge-venom |
xss-vulnerability-scanner | red-team-widow |
state-machine-generator | Charlotte states |
auto-doc-generator | JSDoc README |
PLAN.md — current phase and blockers (Phase IV)../scripts/187web-compiler.sh or .ps1.neuro_toxin block.widow-weaver, swarm-mind, or 187WEBDEV child.{{vars}}.agent-charlotte (research) or silk-sandbox (execution).PLAN.md Session Log.| Mode | Detected when | Default persona | Default prompt |
|---|---|---|---|
high | GPU + AC power or ≥8 cores | ui-spinner | generative-drafting |
low | Battery / NPU-only / <4 cores | edge-venom | edge-ai-deployment-specialist |
standard | Everything else | alpha-architect | ml-systems-architect |
Override: E187WEB_POWER_MODE=high|low|standard
MANIFEST.xml
↓ 187web-compiler
├→ neuro-toxin (inference profile)
├→ widow-weaver (Layer 1–2)
├→ swarm-mind (Layer 3)
├→ 187WEBDEV suite (Layer 4 UI prompts)
├→ agent-charlotte (web crawler, vault relay)
└→ silk-sandbox (safety sentinel)
↓
OmniQube Render Matrix (WebSocket telemetry)
# 187web compiler hook — recompile persona on directory change
_187web_compiler_hook() {
command -v 187web-compiler.sh >/dev/null 2>&1 || return 0
187web-compiler.sh --quiet --write --emit >/dev/null 2>&1 &
}
# Add to chpwd_functions (zsh) or PROMPT_COMMAND (bash)
Manifest is the source of truth. Child skills execute; compiler routes.