Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add jdmorag97-rgb/DDC_Skills_for_AI_Agents_in_Construction --skill toolbox-talk-generatorgit clone --depth 1 https://github.com/jdmorag97-rgb/DDC_Skills_for_AI_Agents_in_ConstructionWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/jdmorag97-rgb/ddc_skills_for_ai_agents_in_construction/toolbox-talk-generator)<a href="https://agentmods.dev/skills/jdmorag97-rgb/ddc_skills_for_ai_agents_in_construction/toolbox-talk-generator"><img src="https://agentmods.dev/badge/skills/jdmorag97-rgb/ddc_skills_for_ai_agents_in_construction/toolbox-talk-generator/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/jdmorag97-rgb/ddc_skills_for_ai_agents_in_construction/toolbox-talk-generator"><img src="https://agentmods.dev/badge/skills/jdmorag97-rgb/ddc_skills_for_ai_agents_in_construction/toolbox-talk-generator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00033 | $0.04136 |
| Opus 5 | $0.00016 | $0.02068 |
| Sonnet 5 | $0.00007 | $0.00827 |
| Haiku 4.5 | $0.00003 | $0.00414 |
Grade A, and why
toolbox-talk-generator scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 8d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
This is a copy
100% identical to toolbox-talk-generator — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 553 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Toolbox Talk Generator
Overview
Automatically generate relevant safety toolbox talks based on daily work activities, weather conditions, recent incidents, and seasonal hazards. Support multiple languages for diverse crews.
"Daily toolbox talks reduce incidents by 30% when relevant to actual work" — DDC Community
How It Works
┌─────────────────────────────────────────────────────────────────┐
│ TOOLBOX TALK GENERATOR │
├─────────────────────────────────────────────────────────────────┤
│ │
│ Inputs Generator Output │
│ ────── ───────── ────── │
│ 📅 Today's work → → 📋 Talk script │
│ 🌤️ Weather → 🤖 AI Engine → 📸 Visual aids │
│ ⚠️ Recent incidents → → ✅ Sign-in sheet │
│ 📆 Season/holiday → → 🌐 Translations │
│ │
└─────────────────────────────────────────────────────────────────┘
Technical Implementation
from dataclasses import dataclass, field
from typing import List, Dict, Optional
from enum import Enum
from datetime import datetime, date
import random
class HazardCategory(Enum):
FALL_PROTECTION = "fall_protection"
ELECTRICAL = "electrical"
EXCAVATION = "excavation"
SCAFFOLDING = "scaffolding"
CRANE_RIGGING = "crane_rigging"
CONFINED_SPACE = "confined_space"
HOT_WORK = "hot_work"
HAZMAT = "hazmat"
HEAT_STRESS = "heat_stress"
COLD_STRESS = "cold_stress"
HOUSEKEEPING = "housekeeping"
PPE = "ppe"
HAND_TOOLS = "hand_tools"
POWER_TOOLS = "power_tools"
MATERIAL_HANDLING = "material_handling"
TRAFFIC = "traffic"
SILICA = "silica"
NOISE = "noise"
@dataclass
class ToolboxTalk:
id: str
title: str
category: HazardCategory
duration_minutes: int
content: Dict[str, str] # sections: intro, hazards, controls, discussion, summary
key_points: List[str]
discussion_questions: List[str]
date_generated: datetime = field(default_factory=datetime.now)
language: str = "en"
@dataclass
class TalkRecord:
talk_id: str
date: datetime
project: str
location: str
presenter: str
attendees: List[str]
topics_covered: List[str]
questions_raised: List[str]
follow_up_actions: List[str]
class ToolboxTalkGenerator:
"""Generate contextual safety toolbox talks."""
# Talk templates by category
TALK_TEMPLATES = {
HazardCategory.FALL_PROTECTION: {
"title": "Fall Protection - Stay Safe at Heights",
"intro": "Falls are the leading cause of death in construction. Today we'll review how to protect ourselves when working at heights.",
"hazards": [
"Unprotected edges and openings",
"Improper ladder use",
"Damaged or missing guardrails",
"Incorrect harness use",
"Unsecured tools and materials"
],
"controls": [
"Always use guardrails at 6 feet or above",
"Inspect harness and lanyards before each use",
"Maintain 3 points of contact on ladders",
"Cover all floor openings",
"Tether tools when working at height"
],
"key_points": [
"100% tie-off required above 6 feet",
"Inspect fall protection daily",
"Know your anchor points",
"Report damaged equipment immediately"
],
"discussion": [
"Where are the fall hazards on our site today?",
"What fall protection will you use?",
"Have you inspected your equipment?"
]
},
HazardCategory.HEAT_STRESS: {
"title": "Heat Stress Prevention",
"intro": "Working in hot conditions can lead to serious illness. Today we'll discuss how to recognize and prevent heat-related illness.",
"hazards": [
"High temperatures and humidity",
"Direct sun exposure",
"Physical exertion",
"Inadequate hydration",
"Lack of acclimatization"
],
"controls": [
"Drink water every 15-20 minutes",
"Take breaks in shade or cool areas",
"Wear light, breathable clothing",
"Know the signs of heat illness",
"Use buddy system to watch each other"
],
"key_points": [
"Water, rest, shade - the three keys",
"Don't wait until you're thirsty to drink",
"Stop work if you feel dizzy or nauseous",
"Acclimatize over 7-14 days"
],
"discussion": [
"Where are the water stations today?",
"Where can you take a cool break?",
"What are the symptoms of heat exhaustion?"
]
},
HazardCategory.ELECTRICAL: {
"title": "Electrical Safety",
"intro": "Electricity can kill instantly. Today we'll review how to work safely around electrical hazards.",
"hazards": [
"Overhead power lines",
"Damaged cords and equipment",
"Wet conditions",
"Missing GFCIs",
"Overloaded circuits"
],
"controls": [
"Maintain 10+ feet from power lines",
"Inspect cords before use",
"Use GFCIs for all power tools",
"Never use damaged equipment",
"Keep electrical away from water"
],
"key_points": [
"Assume all wires are energized",
"Lock out/tag out before work",
"Only qualified personnel do electrical work",
"Report damaged equipment immediately"
],
"discussion": [
"Where are electrical hazards on site today?",
"Are all your tools inspected?",
"Where are the GFCIs located?"
]
},
HazardCategory.HOUSEKEEPING: {
"title": "Good Housekeeping = Safe Workplace",
"intro": "A clean site is a safe site. Poor housekeeping leads to trips, falls, and fires. Let's discuss keeping our work area organized.",
"hazards": [
"Debris and clutter in walkways",
"Improper material storage",
"Tangled cords and hoses",
"Accumulated combustibles",
"Blocked exits and access"
],
"controls": [
"Clean as you go throughout the day",
"Store materials properly",
"Route cords away from walkways",
"Dispose of waste in proper containers",
"Keep exits and aisles clear"
],
"key_points": [
"Clean up immediately after each task",
"Everyone is responsible for housekeeping",
"If you see it, fix it",
"End each day with a clean work area"
],
"discussion": [
"What areas need attention today?",
"Where should materials be stored?",
"Who is responsible for end-of-day cleanup?"
]
},
HazardCategory.SCAFFOLDING: {
"title": "Scaffold Safety",
"intro": "Scaffolds provide safe access for work at height - but only when properly erected and used. Let's review scaffold safety.",
"hazards": [
"Incomplete or damaged scaffolds",
"Missing guardrails or toeboards",
"Overloading",
"Improper access",
"Unstable base"
],
"controls": [
"Only use tagged scaffolds (green tag)",
"Check for complete guardrails",
"Use proper access (ladder, stairs)",
"Don't overload - check capacity",
"Never modify scaffold yourself"
],
"key_points": [
"Green tag = safe to use",
"Red/yellow tag = do not use",
"Inspect before each use",
"Report problems to supervisor"
],
"discussion": [
"Is the scaffold inspected and tagged?",
"What is the load capacity?",
"Where is the proper access point?"
]
}
}
# Weather-related topic mapping
WEATHER_TOPICS = {
"hot": [HazardCategory.HEAT_STRESS],
"cold": [HazardCategory.COLD_STRESS],
"rain": [HazardCategory.ELECTRICAL, HazardCategory.HOUSEKEEPING],
"wind": [HazardCategory.CRANE_RIGGING, HazardCategory.SCAFFOLDING],
"snow": [HazardCategory.COLD_STRESS, HazardCategory.HOUSEKEEPING]
}
# Activity-related topic mapping
ACTIVITY_TOPICS = {
"concrete": [HazardCategory.SILICA, HazardCategory.MATERIAL_HANDLING],
"steel": [HazardCategory.CRANE_RIGGING, HazardCategory.FALL_PROTECTION],
"electrical": [HazardCategory.ELECTRICAL],
"excavation": [HazardCategory.EXCAVATION],
"roofing": [HazardCategory.FALL_PROTECTION, HazardCategory.HEAT_STRESS],
"welding": [HazardCategory.HOT_WORK, HazardCategory.PPE],
"demolition": [HazardCategory.SILICA, HazardCategory.HOUSEKEEPING],
"painting": [HazardCategory.HAZMAT, HazardCategory.PPE]
}
def __init__(self):
self.generated_talks: Dict[str, ToolboxTalk] = {}
self.talk_records: List[TalkRecord] = []
def generate_talk(self, category: HazardCategory,
language: str = "en",
custom_points: List[str] = None) -> ToolboxTalk:
"""Generate toolbox talk for category."""
template = self.TALK_TEMPLATES.get(category)
if not template:
# Generate generic talk
template = self._generate_generic_template(category)
talk_id = f"TBT-{datetime.now().strftime('%Y%m%d%H%M%S')}"
content = {
"intro": template["intro"],
"hazards": "\n".join(f"• {h}" for h in template["hazards"]),
"controls": "\n".join(f"• {c}" for c in template["controls"]),
"summary": f"Remember: {template['key_points'][0]}"
}
key_points = template["key_points"].copy()
if custom_points:
key_points.extend(custom_points)
talk = ToolboxTalk(
id=talk_id,
title=template["title"],
category=category,
duration_minutes=10,
content=content,
key_points=key_points,
discussion_questions=template.get("discussion", []),
language=language
)
self.generated_talks[talk_id] = talk
return talk
def _generate_generic_template(self, category: HazardCategory) -> Dict:
"""Generate generic template for unmapped categories."""
name = category.value.replace("_", " ").title()
return {
"title": f"{name} Safety",
"intro": f"Today we'll discuss {name.lower()} safety and how to protect ourselves.",
"hazards": [
f"Common {name.lower()} hazards on site",
"Lack of awareness",
"Rushing or taking shortcuts",
"Not following procedures"
],
"controls": [
"Follow all safety procedures",
"Use required PPE",
"Report hazards immediately",
"Ask if unsure"
],
"key_points": [
"Safety first - always",
"If unsure, ask your supervisor",
"Report all hazards"
],
"discussion": [
f"What {name.lower()} hazards exist on site today?",
"What controls will you use?",
"Any questions or concerns?"
]
}
def suggest_topics(self, weather: str = None,
activities: List[str] = None,
recent_incidents: List[str] = None) -> List[HazardCategory]:
"""Suggest relevant topics based on context."""
suggestions = set()
# Weather-based suggestions
if weather:
weather_lower = weather.lower()
for condition, topics in self.WEATHER_TOPICS.items():
if condition in weather_lower:
suggestions.update(topics)
# Activity-based suggestions
if activities:
for activity in activities:
activity_lower = activity.lower()
for key, topics in self.ACTIVITY_TOPICS.items():
if key in activity_lower:
suggestions.update(topics)
# Incident-based suggestions
if recent_incidents:
for incident in recent_incidents:
incident_lower = incident.lower()
if "fall" in incident_lower:
suggestions.add(HazardCategory.FALL_PROTECTION)
if "electric" in incident_lower:
suggestions.add(HazardCategory.ELECTRICAL)
if "struck" in incident_lower:
suggestions.add(HazardCategory.MATERIAL_HANDLING)
# Default suggestion if none
if not suggestions:
suggestions.add(HazardCategory.HOUSEKEEPING)
return list(suggestions)
def generate_daily_talk(self, project: str,
weather: str,
activities: List[str],
recent_incidents: List[str] = None) -> ToolboxTalk:
"""Generate contextual daily toolbox talk."""
topics = self.suggest_topics(weather, activities, recent_incidents)
# Pick most relevant topic
primary_topic = topics[0] if topics else HazardCategory.HOUSEKEEPING
# Add context-specific points
custom_points = []
if weather:
custom_points.append(f"Today's weather: {weather} - plan accordingly")
if activities:
custom_points.append(f"Today's focus: {', '.join(activities)}")
talk = self.generate_talk(primary_topic, custom_points=custom_points)
return talk
def format_talk_script(self, talk: ToolboxTalk) -> str:
"""Format talk as presenter script."""
lines = [
f"# {talk.title}",
f"",
f"**Duration:** {talk.duration_minutes} minutes",
f"**Category:** {talk.category.value}",
f"**Date:** {talk.date_generated.strftime('%Y-%m-%d')}",
f"",
f"---",
f"",
f"## Introduction",
f"",
talk.content["intro"],
f"",
f"## Hazards to Watch For",
f"",
talk.content["hazards"],
f"",
f"## How to Protect Yourself",
f"",
talk.content["controls"],
f"",
f"## Key Points to Remember",
f"",
]
for point in talk.key_points:
lines.append(f"✓ {point}")
lines.extend([
f"",
f"## Discussion Questions",
f""
])
for q in talk.discussion_questions:
lines.append(f"❓ {q}")
lines.extend([
f"",
f"---",
f"",
f"**Closing:** Work safe today. If you see something, say something. Any questions?"
])
return "\n".join(lines)
def record_talk(self, talk_id: str, project: str, location: str,
presenter: str, attendees: List[str],
questions: List[str] = None,
follow_ups: List[str] = None) -> TalkRecord:
"""Record completed toolbox talk."""
if talk_id not in self.generated_talks:
raise ValueError(f"Talk {talk_id} not found")
talk = self.generated_talks[talk_id]
record = TalkRecord(
talk_id=talk_id,
date=datetime.now(),
project=project,
location=location,
presenter=presenter,
attendees=attendees,
topics_covered=[talk.category.value],
questions_raised=questions or [],
follow_up_actions=follow_ups or []
)
self.talk_records.append(record)
return record
def get_attendance_summary(self, project: str = None,
start_date: datetime = None) -> Dict:
"""Get toolbox talk attendance summary."""
records = self.talk_records
if project:
records = [r for r in records if r.project == project]
if start_date:
records = [r for r in records if r.date >= start_date]
total_talks = len(records)
total_attendees = sum(len(r.attendees) for r in records)
unique_attendees = len(set(a for r in records for a in r.attendees))
topics_covered = {}
for r in records:
for topic in r.topics_covered:
topics_covered[topic] = topics_covered.get(topic, 0) + 1
return {
"total_talks": total_talks,
"total_attendees": total_attendees,
"unique_workers": unique_attendees,
"avg_attendance": total_attendees / total_talks if total_talks else 0,
"topics_covered": topics_covered
}
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 8d ago First seen · 553 lines · 33 tokens per session scan A 18f8ec84e624
toolbox-talk-generator is a skill published in the GitHub repository jdmorag97-rgb/DDC_Skills_for_AI_Agents_in_Construction (2 stars, last pushed 6mo ago), licensed MIT. It adds 33 tokens to every session and 4,136 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to toolbox-talk-generator, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
ljg-read
Reading companion agent. Accompanies user through any text (books, articles, essays, papers, news) with translation, structural annotation, deep questioning, and cross-domain insights. Detects language, translates English to Chinese (faithfulness-expressiveness-elegance), guides reader to understand the author and…
toolbox-talk-generator
Generate safety toolbox talks for construction crews. Create contextual safety briefings based on weather, work activities, and recent incidents. Support multiple languages.
vietnamese-education-copy
Writes and reviews native-quality Vietnamese (vi-VN) educational, school, academic, and EdTech communication — K-12 report-card remarks and học bạ entries, sổ liên lạc entries and school-to-parent broadcasts, pedagogical nudges, diagnostic pronunciation feedback, adaptive assessment, parental analytics reports…
lov-subtitle-freedom
Create learner-friendly English subtitles with level-aware glosses and optional spoiler-safe subtitle sidecars.
apple-design
Apple's approach to interface design and fluid, physical motion, translated for the web. Use when building or reviewing gesture-driven UI, spring animations, drag/swipe/sheet interactions, momentum and interruptible transitions, translucent materials and depth, typography (optical sizing, tracking, leading)…
stage-and-commit
Stage changed files and create a commit following the repo's translation, formatting, and git workflow rules. Use when the user asks to stage, commit, or both.