一键导入
idumb-governance
Complete iDumb governance protocols - hierarchical delegation, validation patterns, context anchoring, and expert-skeptic mode guidelines
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
Complete iDumb governance protocols - hierarchical delegation, validation patterns, context anchoring, and expert-skeptic mode guidelines
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
Validates iDumb plugin functionality through systematic pivotal trials and stress testing. Ensures all success criteria are met with concrete evidence. Use when testing any iDumb component or before considering implementation complete.
Guides the implementation of the iDumb plugin reboot following the 8-phase pivotal trial methodology. Use when building TypeScript engines, tools, and hooks for intelligent AI governance.
This skill should be used when the user asks to "orchestrate", "delegate", "coordinate agents", "enforce hierarchy", "mindful delegation", "hierarchical thinking", "chain of command", or "orchestrator patterns". Essential for coordinators and high-governance agents to maintain delegation integrity.
Complete iDumb governance protocols - hierarchical delegation, validation patterns, context anchoring, and expert-skeptic mode guidelines
Meta-skill for idumb-builder agent to transform confusing specifications into structured iDumb workflow modules with checkpoints, validation, and integration points. Use when creating reusable workflows, handling overlapping specs, or generating governance modules under .idumb/modules/. This skill enforces the iDumb vision: spec-driven development, context-first validation, drift detection, and hierarchical coordination.
This skill should be used when the user asks to "ingest meta-framework", "transform framework to idumb", "classify framework patterns", "create self-upgrading module", "meta-learning", "framework synthesis", "automated governance enhancement", "pattern extraction", "hierarchical framework analysis", or "iterative framework learning". Essential for idumb-builder to consume external frameworks like BMAD and transform them into iDumb-compatible governance patterns.
| name | idumb-governance |
| description | Complete iDumb governance protocols - hierarchical delegation, validation patterns, context anchoring, and expert-skeptic mode guidelines |
| license | MIT |
| compatibility | opencode |
| metadata | {"audience":"ai-agents","workflow":"governance"} |
Complete protocols for operating within the iDumb hierarchical governance system.
NEVER assume. ALWAYS verify.
Before ANY action:
.idumb/brain/state.jsonEvery validation must return:
result:
status: pass | fail | warning
evidence: [what was checked]
proof: [how it was verified]
timestamp: [when checked]
Purpose: Manage iDumb framework itself
Scope: Restricted to .idumb/ and .opencode/ directories
Agents:
@idumb-supreme-coordinator - Entry-point coordinator (delegation only)@idumb-high-governance - Mid-level coordinator (delegation only)@idumb-meta-builder - ONLY META agent that can write/edit framework files@idumb-meta-validator - Validates framework state (read-only)Rules:
idumb-meta-builder: ONLY agent that can write framework filesidumb-meta-validator: Read-only validationDelegation pattern:
@idumb-high-governance
Task: [what needs doing]
Context: [relevant state]
Requirements: [constraints]
Report: [expected format]
Purpose: Work on user's actual project code
Scope: User project directory (excluding .idumb/, .opencode/)
Agents:
@idumb-project-executor - Executes project code tasks (can write user code)@idumb-project-coordinator - Coordinates project workflows@idumb-project-validator - Validates project code quality@idumb-project-explorer - Explores project codebaseCapabilities:
@idumb-project-executor can write/edit user codetask() toolKey Principle: The distinction is about scope (framework vs user code), not capability or depth. Any agent can delegate to any other agent.
Return format (for validation work):
validation:
check: [what was checked]
status: pass | fail
evidence: [proof]
details: [explanation]
Return format (for execution work):
execution:
action: [what was done]
files: [modified paths]
status: success | partial | failed
changes: [summary]
Check .idumb/ directory integrity:
.idumb/
├── brain/
│ └── state.json # REQUIRED
├── governance/ # Optional
│ └── plugin.log # Created by plugin
└── anchors/ # Optional
Required fields in state.json:
{
"version": "string",
"initialized": "ISO date string",
"framework": "idumb | bmad | planning | custom | none",
"phase": "string",
"lastValidation": "ISO date string | null",
"validationCount": "number",
"anchors": "array",
"history": "array"
}
If planning framework detected:
.planning/ existsROADMAP.md existsPROJECT.md existsCreate anchors for:
| Type | Use | Priority |
|---|---|---|
decision | Strategic choices | critical/high |
context | Background information | normal/high |
checkpoint | Phase completion markers | high |
anchor:
type: decision | context | checkpoint
content: "Brief description of what to remember"
priority: critical | high | normal
.idumb/| File | Purpose | iDumb Access |
|---|---|---|
.planning/ | Planning artifacts | READ ONLY |
ROADMAP.md | Project roadmap | READ ONLY |
PROJECT.md | Project description | READ ONLY |
phases/ | Phase-specific plans | READ ONLY |
After planning operations:
PROJECT.md for current phase.idumb/brain/state.json phaseDuring compaction, the plugin injects:
Before context window fills:
When validation fails:
If state.json is corrupted:
If .idumb/ doesn't exist:
/idumb:init| Command | Purpose | Agent |
|---|---|---|
/idumb:init | Initialize governance | supreme-coordinator |
/idumb:status | Show current state | supreme-coordinator |
/idumb:validate | Run validation | supreme-coordinator |
/idumb:help | Show help | supreme-coordinator |
| Tool | Purpose | Exports |
|---|---|---|
idumb-state | State management | read, write, anchor, history, getAnchors |
idumb-validate | Validation runner | structure, schema, freshness, planningAlignment, default (full) |
idumb-context | Context detection | default (classify), summary, patterns |
Remember: The distinction is META (framework) vs PROJECT (user code), not "levels" or "subagents."