Iot Satellite Forest Fire Early Warning

SKILL.md — IoT Satellite Forest Fire Early Warning Registry

Overview

This document serves as the comprehensive skill registry for the iot-satellite-forest-fire-early-warning system. It defines how skills are registered, resolved, executed, and validated, including complete input/output JSON schemas.

System Architecture

┌─────────────────────────────────────────────────────────────────┐
│                    Skill Orchestrator                            │
│                                                               │
│  ┌─────────────┐  ┌─────────────┐  ┌─────────────┐           │
│  │  Registry   │  │    Hooks    │  │    Tools    │           │
│  └──────┬──────┘  └──────┬──────┘  └──────┬──────┘           │
│         │                │                │                   │
│         └────────────────┼────────────────┘                   │
│                          │                                     │
┌──────────────────────────┼─────────────────────────────────────┐
│                          │                                     │
│                   ┌──────▼──────┐                               │
│                   │  Main Skill │                               │
│                   └──────┬──────┘                               │
│                          │                                     │
│    ┌─────────────────────┼─────────────────────┐              │
│    │                     │                     │               │
┌────▼──────┐  ┌──────────▼──────┐  ┌──────────▼────────┐      │
│ Sub-Skill │  │  Sub-Skill 2   │  │  Sub-Skill N      │      │
│    1      │  │                 │  │                  │      │
└───────────┘  └─────────────────┘  └──────────────────┘      │
│                                                              │
└──────────────────────────────────────────────────────────────┘

Skill Registration

Registration Process

Skills are registered through a multi-step process:

1. Manifest Declaration: Each skill declares itself in SKILL.md frontmatter
2. Schema Validation: Input/output schemas are validated against the registry
3. Hook Binding: Lifecycle hooks are bound to skill events
4. Tool Association: Required tools are associated and validated
5. Quality Gate Assignment: Quality gates are assigned based on skill category

Skill Manifest Schema

{
  "type": "object",
  "required": ["name", "description", "version", "category"],
  "properties": {
    "name": {
      "type": "string",
      "pattern": "^[a-z0-9-]+$",
      "description": "Unique skill identifier in kebab-case"
    },
    "description": {
      "type": "string",
      "minLength": 50,
      "maxLength": 500,
      "description": "Detailed description of when and how to trigger the skill"
    },
    "version": {
      "type": "string",
      "pattern": "^\d+\.\d+\.\d+$",
      "description": "Semantic version"
    },
    "category": {
      "type": "string",
      "enum": ["domain-analysis", "data-processing", "code-generation", "validation"],
      "description": "Skill category for routing"
    },
    "tags": {
      "type": "array",
      "items": {"type": "string"},
      "description": "Tags for skill discovery"
    },
    "compatibility": {
      "type": "object",
      "properties": {
        "required_tools": {"type": "array", "items": {"type": "string"}},
        "python_version": {"type": "string"},
        "dependencies": {"type": "array", "items": {"type": "string"}}
      }
    }
  }
}

Sub-Skill Registry

Available Sub-Skills

| Sub-Skill | Purpose | Input Schema | Output Schema | |-----------|---------|--------------|---------------| | sub-gather-requirements | Intake and requirements clarification | RequirementsInput | RequirementsOutput | | sub-evidence-collector | Data fetching and evidence collection | EvidenceInput | EvidenceOutput | | sub-core-analysis | Domain analysis and computation | AnalysisInput | AnalysisOutput | | sub-knowledge-updater | Knowledge base querying | KnowledgeQueryInput | KnowledgeQueryOutput | | sub-advisor | Synthesis and recommendation | AdvisorInput | AdvisorOutput |

Input/Output Schemas

RequirementsInput

{
  "type": "object",
  "properties": {
    "user_message": {
      "type": "string",
      "description": "Raw user input message"
    },
    "language": {
      "type": "string",
      "enum": ["en", "vi"],
      "description": "User's preferred language"
    },
    "context": {
      "type": "object",
      "description": "Additional context from previous interactions"
    }
  },
  "required": ["user_message"]
}

RequirementsOutput

{
  "type": "object",
  "properties": {
    "object_of_analysis": {
      "type": "string",
      "description": "What is being analyzed"
    },
    "scope": {
      "type": "string",
      "description": "Analysis scope and boundaries"
    },
    "timeframe": {
      "type": "string",
      "description": "Analysis timeframe"
    },
    "available_inputs": {
      "type": "array",
      "items": {"type": "string"},
      "description": "Available data inputs"
    },
    "target_audience": {
      "type": "string",
      "description": "Target audience for output"
    },
    "language": {
      "type": "string",
      "enum": ["en", "vi"]
    },
    "analysis_type": {
      "type": "string",
      "description": "Type of analysis required"
    },
    "valid": {
      "type": "boolean",
      "description": "Whether requirements are valid"
    }
  },
  "required": ["object_of_analysis", "language", "valid"]
}

EvidenceInput

{
  "type": "object",
  "properties": {
    "requirements": {"$ref": "#/definitions/RequirementsOutput"},
    "data_sources": {
      "type": "array",
      "items": {"type": "string"},
      "description": "Preferred data sources"
    }
  },
  "required": ["requirements"]
}

EvidenceOutput

{
  "type": "object",
  "properties": {
    "current_data": {
      "type": "object",
      "description": "Current real-time data with source and timestamp"
    },
    "authoritative_docs": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "title": {"type": "string"},
          "source": {"type": "string"},
          "url": {"type": "string"},
          "date": {"type": "string"},
          "tier": {"type": "string", "enum": ["Tier 1", "Tier 2", "Tier 3", "Tier 4"]}
        }
      },
      "description": "Authoritative documents and standards"
    },
    "recent_news": {
      "type": "array",
      "items": {"type": "object"},
      "description": "Recent developments from domain sources"
    },
    "reference_benchmarks": {
      "type": "object",
      "description": "Reference benchmark data"
    },
    "degradation_level": {
      "type": "integer",
      "minimum": 0,
      "maximum": 4,
      "description": "Data availability degradation level"
    }
  },
  "required": ["current_data", "degradation_level"]
}

AnalysisInput

{
  "type": "object",
  "properties": {
    "requirements": {"$ref": "#/definitions/RequirementsOutput"},
    "evidence": {"$ref": "#/definitions/EvidenceOutput"},
    "parameters": {
      "type": "object",
      "properties": {
        "forest_region": {"type": "string"},
        "weather_data": {
          "type": "object",
          "properties": {
            "temperature": {"type": "number"},
            "humidity": {"type": "number"},
            "wind_speed": {"type": "number"},
            "rainfall": {"type": "number"}
          }
        },
        "sensor_config": {
          "type": "object",
          "description": "IoT sensor configuration"
        },
        "satellite_coverage": {
          "type": "boolean",
          "description": "Satellite data availability"
        }
      }
    }
  },
  "required": ["requirements", "evidence"]
}

AnalysisOutput

{
  "type": "object",
  "properties": {
    "fire_danger_index": {
      "type": "object",
      "properties": {
        "name": {"type": "string", "enum": ["FWI", "NFDRS"]},
        "value": {"type": "number"},
        "components": {"type": "object"},
        "danger_class": {"type": "string"}
      }
    },
    "iot_deployment": {
      "type": "object",
      "properties": {
        "sensors": {"type": "array", "items": {"type": "object"}},
        "coverage_area": {"type": "string"},
        "data_frequency": {"type": "string"}
      }
    },
    "satellite_monitoring": {
      "type": "object",
      "properties": {
        "active_fires": {"type": "array", "items": {"type": "object"}},
        "vegetation_dryness": {"type": "object"},
        "refresh_interval": {"type": "string"}
      }
    },
    "alerts": {
      "type": "object",
      "properties": {
        "thresholds": {"type": "object"},
        "notification_channels": {"type": "array"}
      }
    },
    "uav_confirmation": {
      "type": "object",
      "properties": {
        "deployment_criteria": {"type": "object"},
        "confirmation_protocol": {"type": "string"}
      }
    },
    "scenarios": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "name": {"type": "string"},
          "probability": {"type": "number"},
          "impact": {"type": "string"},
          "actions": {"type": "array"}
        }
      }
    }
  },
  "required": ["fire_danger_index", "scenarios"]
}

KnowledgeQueryInput

{
  "type": "object",
  "properties": {
    "keywords": {
      "type": "array",
      "items": {"type": "string"},
      "minItems": 1
    },
    "max_results": {
      "type": "integer",
      "default": 5,
      "minimum": 1,
      "maximum": 20
    },
    "required_tiers": {
      "type": "array",
      "items": {"type": "string", "enum": ["Tier 1", "Tier 2", "Tier 3", "Tier 4"]}
    }
  },
  "required": ["keywords"]
}

KnowledgeQueryOutput

{
  "type": "object",
  "properties": {
    "evidence": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "title": {"type": "string"},
          "authors": {"type": "array", "items": {"type": "string"}},
          "year": {"type": "integer"},
          "doi": {"type": "string"},
          "journal": {"type": "string"},
          "tier": {"type": "string", "enum": ["Tier 1", "Tier 2", "Tier 3", "Tier 4"]},
          "relevance_score": {"type": "number"},
          "citation_count": {"type": "integer"}
        }
      }
    },
    "coverage_rating": {
      "type": "string",
      "enum": ["comprehensive", "adequate", "limited", "insufficient"]
    },
    "gaps": {
      "type": "array",
      "items": {"type": "string"},
      "description": "Identified knowledge gaps for crawl pipeline"
    }
  },
  "required": ["evidence", "coverage_rating"]
}

AdvisorInput

{
  "type": "object",
  "properties": {
    "requirements": {"$ref": "#/definitions/RequirementsOutput"},
    "evidence": {"$ref": "#/definitions/EvidenceOutput"},
    "analysis": {"$ref": "#/definitions/AnalysisOutput"},
    "knowledge": {"$ref": "#/definitions/KnowledgeQueryOutput"}
  },
  "required": ["requirements", "analysis"]
}

AdvisorOutput

{
  "type": "object",
  "properties": {
    "verdict": {
      "type": "string",
      "enum": ["Low Fire Risk", "Monitor (elevated)", "Critical Alert", "Dispatch", "Inconclusive"]
    },
    "confidence": {
      "type": "string",
      "enum": ["high", "medium", "low"]
    },
    "scenarios": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "name": {"type": "string"},
          "probability": {"type": "number"},
          "impact": {"type": "string"},
          "actions": {"type": "array"}
        }
      }
    },
    "key_risks": {
      "type": "array",
      "items": {"type": "string"}
    },
    "evidence_chain": {
      "type": "array",
      "items": {
        "type": "object",
        "properties": {
          "claim": {"type": "string"},
          "source": {"type": "string"},
          "tier": {"type": "string"}
        }
      }
    },
    "remediation": {
      "type": "array",
      "items": {"type": "object"}
    },
    "disclosure": {
      "type": "string",
      "description": "Mandatory disclosure before recommendation"
    }
  },
  "required": ["verdict", "disclosure"]
}

Skill Resolution

Resolution Algorithm

Skills are resolved through a multi-stage process:

1. Trigger Detection: User input is analyzed for skill trigger patterns
2. Candidate Matching: Skills are scored based on description similarity
3. Compatibility Check: Required tools and dependencies are verified
4. Context Validation: Input context is validated against skill schemas
5. Skill Selection: Highest scoring compatible skill is selected

Trigger Patterns

| Pattern Type | Example | Weight | |--------------|---------|--------| | Direct invocation | /iot-satellite-forest-fire-early-warning | 1.0 | | Domain keywords | "forest fire risk", "fire danger index" | 0.8 | | Task description | "design early warning system" | 0.7 | | Contextual | analyze vegetation, IoT sensors | 0.6 |

Skill Execution

Execution Protocol

1. PRE-FLIGHT CHECKS
   ├─ Validate configuration
   ├─ Check tool availability
   ├─ Load skill metadata
   └─ Initialize execution context

2. STEP EXECUTION
   For each step in harness:
   ├─ Emit BEFORE_STEP_EXECUTE hook
   ├─ Validate step input against schema
   ├─ Execute step logic
   ├─ Validate step output against schema
   ├─ Emit AFTER_STEP_EXECUTE hook
   └─ Handle errors with graceful degradation

3. QUALITY GATE VALIDATION
   For each quality gate:
   ├─ Emit BEFORE_QUALITY_GATE hook
   ├─ Validate against gate criteria
   ├─ Attempt auto-fix if failed
   ├─ Retry up to max_attempts
   ├─ Emit AFTER_QUALITY_GATE hook
   └─ Flag limitation if still failed

4. OUTPUT FORMATTING
   ├─ Format according to output template
   ├─ Apply language translation
   ├─ Validate all sections present
   └─ Emit AFTER_SKILL_EXECUTE hook

Error Handling

Errors are handled through a structured recovery system:

| Error Level | Behavior | Recovery | |-------------|----------|----------| | WARNING | Log and continue | Use fallback value | | ERROR | Retry with backoff | Alternate source/method | | CRITICAL | Flag and continue | Graceful degradation | | FATAL | Abort execution | Emit error output |

Quality Gates

Universal Gates (U1-U6)

| Gate | Criteria | Auto-Fix | |------|----------|----------| | U1 | ≥3 sources, ≥1 academic/authoritative | Fetch from knowledge base | | U2 | Disclosure before recommendation | Prepend disclosure | | U3 | Evidence hierarchy per source | Annotate tiers | | U4 | Language matches preference | Translate | | U5 | Output template complete | Reformat | | U6 | Claims traceable to sources | Flag or annotate |

Domain Gates (G1-G4)

| Gate | Criteria | Auto-Fix | |------|----------|----------| | G1 | Fire-danger index computed | Compute index | | G2 | IoT sensors deployed | Add deployment plan | | G3 | Satellite monitoring | Add monitoring config | | G4 | Alert thresholds & UAV | Add alert/UAV config |

Lifecycle Hooks

Hook Execution Order

Hooks are executed in priority order:

1. CRITICAL (0): Security, validation, critical checks
2. HIGH (50): Core functionality
3. NORMAL (100): Standard hooks
4. LOW (150): Optional enhancements
5. MONITORING (200): Logging, metrics, analytics

Available Hooks

See hooks/lifecycle.py for complete hook definitions and registration API.

Tool Invocation

Tool Registration

Tools are registered through the @tool decorator:

@tool(
    name="compute_fwi",
    description="Compute Fire Weather Index",
    category=ToolCategory.ANALYSIS,
)
def compute_fwi(temperature: float, humidity: float, wind_sp

…

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [dungnotnull](https://github.com/dungnotnull)
- **Source:** [dungnotnull/iot-satellite-forest-fire-early-warning-agent-skill](https://github.com/dungnotnull/iot-satellite-forest-fire-early-warning-agent-skill)
- **License:** MIT

Install and usage instructions live in the source repository linked above.