Agent Directory Record Example
Skill Tags (Taxonomy)
├── analytical_skills
│ ├── analytical_skills
│ ├── coding_skills
│ │ ├── code_optimization
│ │ ├── code_templates
│ │ ├── code_to_docstrings
│ │ ├── coding_skills
│ │ └── text_to_code
│ └── mathematical_reasoning
│ ├── geometry
│ ├── math_word_problems
│ ├── mathematical_reasoning
│ ├── pure_math_operations
│ └── theorem_proving
├── audio
│ ├── audio_classification
│ ├── audio_to_audio
│ └── audio
├── base_skill
├── images_computer_vision
│ ├── depth_estimation
│ ├── image_classification
│ ├── image_feature_extraction
│ ├── image_generation
│ ├── image_segmentation
│ ├── image_to_3d
│ ├── image_to_image
│ ├── images_computer_vision
│ ├── keypoint_detection
│ ├── mask_generation
│ ├── object_detection
│ └── video_classification
├── multi_modal
│ ├── any_to_any
│ ├── audio_processing
│ │ ├── audio_processing
│ │ ├── speech_recognition
│ │ └── text_to_speech
│ ├── image_processing
│ │ ├── image_processing
│ │ ├── image_to_text
│ │ ├── text_to_3d
│ │ ├── text_to_image
│ │ ├── text_to_video
│ │ └── visual_qa
│ └── multi_modal
├── nlp
│ ├── analytical_reasoning
│ │ ├── analytical_reasoning
│ │ ├── fact_verification
│ │ ├── inference_deduction
│ │ └── problem_solving
│ ├── creative_content
│ │ ├── creative_content
│ │ ├── poetry_writing
│ │ └── storytelling
│ ├── ethical_interaction
│ │ ├── bias_mitigation
│ │ ├── content_moderation
│ │ └── ethical_interaction
│ ├── feature_extraction
│ │ ├── feature_extraction
│ │ └── model_feature_extraction
│ ├── information_retrieval_synthesis
│ │ ├── document_passage_retrieval
│ │ ├── fact_extraction
│ │ ├── information_retrieval_synthesis
│ │ ├── knowledge_synthesis
│ │ ├── question_answering
│ │ ├── search
│ │ └── sentence_similarity
│ ├── language_translation
│ │ ├── language_translation
│ │ ├── multilingual_understanding
│ │ └── translation
│ ├── natural_language_generation
│ │ ├── dialogue_generation
│ │ ├── nlg
│ │ ├── paraphrasing
│ │ ├── question_generation
│ │ ├── story_generation
│ │ ├── style_transfer
│ │ ├── summarization
│ │ └── text_completion
│ ├── natural_language_understanding
│ │ ├── contextual_comprehension
│ │ ├── entity_recognition
│ │ ├── nlu
│ │ └── semantic_understanding
│ ├── nlp
│ ├── personalization
│ │ ├── personalization
│ │ ├── style_adjustment
│ │ └── user_adaptation
│ ├── text_classification
│ │ ├── natural_language_inference
│ │ ├── sentiment_analysis
│ │ ├── text_classification
│ │ └── topic_labeling
│ └── token_classification
│ ├── named_entity_recognition
│ ├── pos_tagging
│ └── token_classification
├── retrieval_augmented_generation
│ ├── document_or_database_question_answering
│ ├── generation_of_any
│ ├── retrieval_augmented_generation
│ └── retrieval_of_information
│ ├── document_retrieval
│ ├── indexing
│ ├── retrieval_of_information
│ └── search
└── tabular_text
├── tabular_classification
├── tabular_regression
└── tabular_text
Record Examples without Digests (Content Identifier)
Email Reviewer AI Agent
{
"name": "agntcy/email_reviewer",
"skills": [
{
"class_uid": 10101,
"class_name": "Contextual Comprehension",
"category_uid": 1,
"category_name": "Natural Language Processing"
},
{
"class_uid": 10206,
"class_name": "Text Style Transfer",
"category_uid": 1,
"category_name": "Natural Language Processing"
},
{
"class_uid": 10602,
"class_name": "Tone and Style Adjustment",
"category_uid": 1,
"category_name": "Natural Language Processing"
},
{
"class_uid": 10702,
"class_name": "Problem Solving",
"category_uid": 1,
"category_name": "Natural Language Processing"
},
{
"class_uid": 10203,
"class_name": "Text Paraphrasing",
"category_uid": 1,
"category_name": "Natural Language Processing"
},
{
"class_uid": 10303,
"class_name": "Knowledge Synthesis",
"category_uid": 1,
"category_name": "Natural Language Processing"
}
],
"authors": [
"Cisco Systems Inc."
],
"version": "v1.0.0",
"locators": [
{
"url": "https://github.com/agntcy/agentic-apps/tree/main/email_reviewer",
"type": "source-code"
},
{
"url": "https://github.com/agntcy/agentic-apps/tree/main/email_reviewer/pyproject.toml",
"type": "python-package"
}
],
"signature": {
"algorithm": "SHA2_256",
"signature": "MEUCIQCIlAthHnRAOeHqVqVvy/KW2xej6nTPdsnpmSmHyDoExAIgbl2j/2dFr6oGAdUJG[...]",
"signed_at": "2025-05-21T16:43:35+02:00",
"certificate": "MIICzjCCAlWgAwIBAgIUfTQ0MzM0WhI[...]",
"content_type": "application/vnd.dev.sigstore.bundle.v0.3+json",
"content_bundle": "eyJtZWRpYVR5cGUiOiJhcHBsaWNhdGlvbi92bm0lCQWd [...]"
},
"created_at": "2025-04-24T12:00:00Z",
"extensions": [
{
"data": {
"sbom": {
"name": "email_reviewer",
"packages": [
{
"name": "dotenv",
"version": "^0.9.9"
},
{
"name": "llama-index-core",
"version": "^0.12.30"
},
{
"name": "llama-index-llms-azure-openai",
"version": "^0.3.1"
},
{
"name": "agntcy_acp",
"version": "v0.1.0a2"
}
]
}
},
"name": "schema.oasf.agntcy.org/features/runtime/framework",
"version": "v0.0.0"
},
{
"data": {
"type": "python",
"version": ">=3.9,<4.0"
},
"name": "schema.oasf.agntcy.org/features/runtime/language",
"version": "v0.0.0"
}
],
"annotations": {
"type": "llama-index"
},
"description": "Agent in charge of reviewing and correcting emails addressed to a specific audience.",
"schema_version": "v0.3.1"
}
The content identifier of the record is a SHA-256 hash digests which makes it
- Globally unique
- Content-addressable
- Collision-resistant
- Immutable