Skip to main content

content-flagging

Establishes a comprehensive framework for evaluating, classifying, and disposing of flagged content within the platform's content moderation ecosystem.

الانتقال إلى التثبيت

معلومات المصدر

المستودع
vivekhaldar/proceda
آخر نشاط في المصدر
٢٢ مارس ٢٠٢٦ في ٠٥:٥٩
لغة SKILL.md المكتشفة
الإنجليزية
النجوم
١٦
التفرعات
٢

خيارات التثبيت

يُحدَّد Prompt الذي يراجع المصدر أولًا بشكل افتراضي. يمكنك التبديل إلى أمر مباشر أو تنزيل نسخة محلية.

مراجعة ملفات المصدر

اقرأ SKILL.md وأي ملفات مرافقة يعرضها SkillsMP قبل أن تقرر التثبيت.

مستكشف الملفات
2 ملفات

عرض SKILL.md

SKILL.md
تعليمات المصدر · معاينة للقراءة فقط
name
content-flagging
description
Establishes a comprehensive framework for evaluating, classifying, and disposing of flagged content within the platform's content moderation ecosystem.
required_tools
["calculateBotProbabilityIndex","calculate_user_trust_score","calculateContentSeverityIndex","determineFinalDecision"]
### Step 1: Calculate Bot Probability Index Utilize the `calculateBotProbabilityIndex` tool to assess the likelihood of automated activity, incorporating device consistency validation. This tool will take the following inputs: - `userid` from content metadata - `is_possible_bot` (initial bot assessment) - `Captcha_tries` (number of captcha attempts) - `device_type` from device information - `os` (operating system) from device information - `browser` (browser specification) from device information The outputs will be the `bot_probability_index` and the `device_consistency_score`. ### Step 2: Calculate User Trust Score Compute the user's overall trust coefficient using the `calculate_user_trust_score` tool, factoring in historical behavior, geographic risk, and device consistency. This tool requires: - `userid` from content metadata - `NumberofPreviousPosts` (user's total post count) - `CountofFlaggedPosts` (count of user's previously flagged posts) - `Latitude` and `Longitude` from content metadata (geolocation coordinates) - `bot_probability_index` obtained from Step 1 - `device_consistency_score` obtained from Step 1 The output will be the `user_trust_score`. ### Step 3: Assess Content Severity Determine the severity of the content violation using the `calculateContentSeverityIndex` tool, integrating primary and secondary violation analyses. Provide the following inputs: - `content_id` from content metadata - `PrimaryViolationType` from violation data - `SecondaryViolationType` from violation data (if applicable) - `PrimaryViolation_Confidence` from violation data - `SecondaryViolation_Confidence` from violation data (if applicable) The output will be the `content_severity_index`. ### Step 4: Determine Final Content Disposition Utilize the `determineFinalDecision` tool to combine all moderation metrics and establish the final action for the flagged content. This tool needs: - `content_id` from content metadata - `user_trust_score` obtained from Step 2 - `content_severity_index` obtained from Step 3 - `bot_probability_index` obtained from Step 1 - `NumberofPreviousPosts` (user's total post count) - `CountofFlaggedPosts` (count of user's previously flagged posts) The tool will output the final disposition (e.g., `removed`, `warning`, `user_banned`, `allowed`). [APPROVAL REQUIRED]
عرض على GitHub