{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"github-louisshark--chatgpt_system_prompt","slug":"louisshark--chatgpt_system_prompt","name":"chatgpt_system_prompt","type":"prompt","url":"https://github.com/LouisShark/chatgpt_system_prompt","page_url":"https://unfragile.ai/louisshark--chatgpt_system_prompt","categories":["prompt-engineering"],"tags":["gpt","prompt","prompt-engineering"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"github-louisshark--chatgpt_system_prompt__cap_0","uri":"capability://automation.workflow.automated.toc.generation.for.prompt.collections","name":"automated-toc-generation-for-prompt-collections","description":"Automatically generates and maintains table of contents (TOC) files across the repository using a GitHub Actions workflow that triggers on main branch pushes and PR merges. The system uses Python scripts (idxtool.py, gptparser.py) to enumerate prompt files, parse their metadata, and rebuild TOC.md files in the root and all subdirectories under /prompts/, ensuring navigation links remain current as new prompts are added or modified without manual intervention.","intents":["I need to keep my prompt collection indexed and navigable as I add hundreds of new prompts","I want contributors to submit new prompts without manually updating table of contents","I need to generate hierarchical navigation for 1,100+ custom GPTs organized by category"],"best_for":["open-source prompt repositories with frequent community contributions","teams managing large collections of AI system prompts across multiple categories","documentation maintainers who want to avoid manual TOC updates"],"limitations":["Requires GitHub Actions CI/CD integration — not portable to other version control systems without modification","TOC generation latency increases linearly with repository size (1,100+ GPTs may take 10-30 seconds per rebuild)","No incremental indexing — rebuilds entire TOC on every trigger rather than delta updates","Markdown-only output format — cannot generate alternative documentation formats (HTML, JSON indices)"],"requires":["GitHub repository with Actions enabled","Python 3.6+ for idxtool.py and gptparser.py scripts",".github/workflows/build-toc.yaml configuration file","Standardized markdown format for prompt files with parseable metadata headers"],"input_types":["markdown files with YAML/frontmatter metadata","directory structure under /prompts/ with subdirectories (gpts, official-product, opensource-prj)"],"output_types":["markdown TOC files (TOC.md)","hierarchical navigation links with file paths"],"categories":["automation-workflow","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-louisshark--chatgpt_system_prompt__cap_1","uri":"capability://data.processing.analysis.prompt.metadata.parsing.and.standardization","name":"prompt-metadata-parsing-and-standardization","description":"Parses markdown prompt files using gptparser.py to extract and standardize metadata fields (name, description, author, tags, etc.) from YAML frontmatter and markdown headers. The parser maintains a dictionary of supported fields with display names and processing order, enabling consistent formatting across heterogeneous prompt sources (official OpenAI/Anthropic products, community GPTs, open-source projects) and enabling downstream indexing and search capabilities.","intents":["I need to extract structured metadata from unstructured prompt markdown files to build searchable indices","I want to normalize prompt metadata across different sources (ChatGPT, Claude, Grok) that use different documentation conventions","I need to validate that contributed prompts follow the repository's metadata schema before merging"],"best_for":["prompt repository maintainers aggregating prompts from multiple vendors with inconsistent formats","teams building searchable prompt databases or discovery tools","developers creating prompt validation pipelines for CI/CD workflows"],"limitations":["Markdown-only parsing — cannot extract metadata from JSON, YAML, or other structured formats without additional parsers","Fragile to formatting variations — prompts with non-standard YAML frontmatter or missing headers may fail to parse completely","No semantic understanding of prompt content — extracts only syntactic metadata fields, not capability descriptions or use-case inference","Single-pass parsing — no error recovery or fallback strategies for malformed metadata"],"requires":["Python 3.6+ with standard library (no external dependencies mentioned)","Markdown files with YAML frontmatter or standardized header format","Consistent field naming conventions across all prompt files"],"input_types":["markdown files (.md) with YAML frontmatter","markdown headers and structured text sections"],"output_types":["parsed metadata dictionary with standardized field names","validation results indicating missing or malformed fields"],"categories":["data-processing-analysis","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-louisshark--chatgpt_system_prompt__cap_10","uri":"capability://memory.knowledge.custom.gpt.and.development.ide.assistant.patterns","name":"custom-gpt-and-development-ide-assistant-patterns","description":"Documents patterns and system prompts for custom GPTs and development IDE assistants (including Grimoire Coding Assistant and other specialized tools) organized in /prompts/gpts/. The collection includes 1,100+ examples of how developers structure prompts for specific domains (coding, finance, education, etc.), providing a comprehensive reference for understanding custom GPT design patterns and specialized assistant architectures.","intents":["I want to study how successful custom GPTs structure their system prompts for specific domains","I need examples of coding assistant prompts to understand best practices for developer tools","I want to learn from the Grimoire Coding Assistant and other specialized IDE assistants"],"best_for":["developers building custom GPTs for specialized domains","teams designing AI-powered IDE assistants and developer tools","prompt engineers studying domain-specific prompt patterns","researchers analyzing the diversity of custom GPT use cases"],"limitations":["No quality filtering — all 1,100+ GPTs are included regardless of effectiveness or adoption","No usage metrics — cannot identify which patterns are most successful or widely used","No domain-specific organization beyond directory structure — no semantic tagging or capability-based search","Snapshot in time — does not reflect real-time updates to custom GPTs or their popularity"],"requires":["GitHub repository access to /prompts/gpts/ directory","Understanding of custom GPT architecture and system prompt design","Ability to browse and compare multiple prompt examples"],"input_types":["custom GPT system prompts","domain-specific prompt patterns (coding, finance, education, etc.)","IDE assistant and specialized tool prompts"],"output_types":["markdown files containing custom GPT prompts","organized examples by domain and use case","pattern references for building similar GPTs"],"categories":["memory-knowledge","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-louisshark--chatgpt_system_prompt__cap_2","uri":"capability://memory.knowledge.multi.source.prompt.aggregation.and.curation","name":"multi-source-prompt-aggregation-and-curation","description":"Aggregates and organizes system prompts from three distinct sources (official-product: ChatGPT/Claude/Grok, gpts: 1,100+ community-created custom GPTs, opensource-prj: open-source AI projects) into a unified repository structure with separate TOC hierarchies. The architecture uses directory-based organization (/prompts/gpts/, /prompts/official-product/, /prompts/opensource-prj/) to maintain source separation while enabling cross-source discovery and comparison through unified indexing.","intents":["I want to study how different AI vendors (OpenAI, Anthropic, xAI) structure their system prompts to understand best practices","I need to find and compare community-created GPTs across different domains (coding, finance, education) in one place","I want to contribute my own custom GPT prompt to a curated collection without forking the entire repository"],"best_for":["prompt engineers and researchers studying AI system prompt design patterns","developers building custom GPTs who want to learn from existing examples","teams evaluating different AI vendors' instruction-following approaches","open-source contributors wanting to share prompt innovations"],"limitations":["No deduplication across sources — identical or near-identical prompts from different vendors are stored separately, increasing repository size","No version control for prompt evolution — historical changes to official product prompts are not tracked","Limited metadata about prompt provenance — no timestamps, source URLs, or licensing information for community GPTs","No quality curation — all submitted prompts are included without evaluation of effectiveness or safety"],"requires":["GitHub repository with write access for contributions","Understanding of repository structure and contribution guidelines (CONTRIBUTING.md)","Markdown files following the standardized prompt format"],"input_types":["system prompts from OpenAI ChatGPT, Anthropic Claude, xAI Grok","custom GPT definitions with metadata","open-source AI project prompts"],"output_types":["organized markdown files in source-specific directories","hierarchical TOC files enabling navigation by source and category"],"categories":["memory-knowledge","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-louisshark--chatgpt_system_prompt__cap_3","uri":"capability://safety.moderation.prompt.injection.and.jailbreak.technique.documentation","name":"prompt-injection-and-jailbreak-technique-documentation","description":"Documents and catalogs prompt injection techniques, jailbreak methods, and prompt leaking knowledge as a research and educational resource. The repository includes specific files like GrokJailbreakPrompt.md and security-focused documentation (SECURITY.md) that explain how system prompts can be extracted, bypassed, or manipulated, serving as both a learning resource and a reference for understanding AI safety vulnerabilities.","intents":["I need to understand common prompt injection and jailbreak techniques to build safer AI systems","I want to study real-world examples of prompt leaking to improve my system prompt design","I need to document known vulnerabilities in AI systems for security research and red-teaming"],"best_for":["AI safety researchers and red-teamers evaluating system prompt robustness","security engineers building guardrails against prompt injection attacks","developers designing custom GPTs who want to understand attack vectors","academic researchers studying AI alignment and instruction-following vulnerabilities"],"limitations":["Educational content only — does not provide automated detection or mitigation of prompt injection attacks","Techniques may become outdated as AI vendors patch vulnerabilities or improve instruction-following","No systematic taxonomy of injection techniques — documentation is ad-hoc rather than comprehensively categorized","Potential misuse risk — detailed jailbreak techniques could enable malicious actors to bypass safety measures"],"requires":["Understanding of AI system prompts and instruction-following mechanisms","Familiarity with prompt engineering concepts","Ethical commitment to responsible disclosure and security research"],"input_types":["markdown documentation of jailbreak techniques","example prompts demonstrating injection vulnerabilities","security guidelines and best practices"],"output_types":["educational markdown files explaining attack vectors","example prompts showing vulnerability demonstrations","security recommendations and mitigation strategies"],"categories":["safety-moderation","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-louisshark--chatgpt_system_prompt__cap_4","uri":"capability://search.retrieval.custom.gpt.discovery.and.browsing","name":"custom-gpt-discovery-and-browsing","description":"Enables discovery and browsing of 1,100+ community-created custom GPTs through hierarchical organization by category (coding, finance, education, etc.) with automated TOC generation and file enumeration. The enum_gpts() and find_gptfile() functions in idxtool.py support both directory-based browsing and ID/URL-based lookup, allowing users to search for GPTs by name, category, or functionality without requiring a database backend.","intents":["I want to find existing custom GPTs in a specific domain (e.g., quantitative finance) to understand how others structure them","I need to browse all available GPTs organized by category to discover new tools and use cases","I want to look up a specific GPT by its ID or URL to view its system prompt and configuration"],"best_for":["developers and prompt engineers exploring custom GPT design patterns","non-technical users discovering useful GPTs for their workflows","teams evaluating community-created solutions before building in-house alternatives","researchers studying the diversity of custom GPT use cases"],"limitations":["No full-text search — discovery is limited to directory browsing and file enumeration, not semantic search across prompt content","No user ratings or popularity metrics — all GPTs are presented equally regardless of quality or adoption","No filtering by capability or use case — users must manually browse categories to find relevant GPTs","No real-time updates — GPT discovery reflects repository state at last commit, not live custom GPT marketplace data"],"requires":["GitHub repository access (read-only for browsing)","Understanding of directory structure (/prompts/gpts/)","Markdown viewer or text editor to read GPT files"],"input_types":["GPT file names and directory structure","GPT metadata (ID, name, description)"],"output_types":["list of available GPTs organized by category","markdown files containing GPT system prompts and configuration","file paths and URLs for direct access"],"categories":["search-retrieval","memory-knowledge"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-louisshark--chatgpt_system_prompt__cap_5","uri":"capability://memory.knowledge.system.prompt.comparison.across.vendors","name":"system-prompt-comparison-across-vendors","description":"Enables side-by-side comparison of system prompts from different AI vendors (OpenAI ChatGPT, Anthropic Claude, xAI Grok, Google AI tools) by organizing official product prompts in /prompts/official-product/ with vendor-specific subdirectories. Users can examine how different vendors structure instructions, handle edge cases, and implement safety guidelines by reading and comparing prompts like ChatGPT system.md, Claude Code System, and Grok2.md/Grok3.md files.","intents":["I want to understand how OpenAI, Anthropic, and xAI differ in their system prompt design and instruction-following approaches","I need to study vendor-specific safety guidelines and constraint handling to inform my own system prompt design","I want to analyze how different vendors structure prompts for specialized tasks (code generation, research, deep search)"],"best_for":["prompt engineers studying best practices across different AI vendors","researchers analyzing AI instruction-following and alignment approaches","teams evaluating which vendor's approach best fits their use case","developers building multi-vendor AI applications who need to understand prompt differences"],"limitations":["Static snapshots only — prompts are not automatically updated when vendors release new versions, requiring manual updates","No version history — cannot track how vendor prompts evolve over time or compare across versions","Incomplete coverage — not all vendor products are documented (e.g., missing some Google AI tools)","No semantic analysis — repository provides raw prompts without comparative analysis or structured insights"],"requires":["GitHub repository access to /prompts/official-product/ directory","Understanding of different vendors' product lines (ChatGPT, Claude, Grok, etc.)","Ability to read and interpret system prompts"],"input_types":["official system prompts from OpenAI, Anthropic, xAI, Google","vendor-specific prompt variations (e.g., Grok2 vs Grok3, Grok3WithDeepSearch)"],"output_types":["markdown files containing vendor system prompts","organized directory structure enabling vendor-by-vendor comparison"],"categories":["memory-knowledge","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-louisshark--chatgpt_system_prompt__cap_6","uri":"capability://automation.workflow.contribution.workflow.and.validation.guidelines","name":"contribution-workflow-and-validation-guidelines","description":"Provides structured contribution guidelines (CONTRIBUTING.md) and security policies (SECURITY.md) that define how community members can submit new prompts, validate metadata, and ensure quality standards. The workflow integrates with GitHub's pull request system and automated TOC generation, enabling contributors to add new prompts without manually updating indices while maintaining repository integrity through validation checks.","intents":["I want to contribute my custom GPT prompt to the repository without manually updating table of contents","I need to understand what metadata and format are required for my prompt submission to be accepted","I want to report a security vulnerability or jailbreak technique responsibly without enabling malicious use"],"best_for":["open-source contributors wanting to share prompt innovations with the community","prompt engineers building custom GPTs who want their work to be discoverable","security researchers responsibly disclosing AI vulnerabilities","repository maintainers enforcing quality standards and preventing spam/malicious contributions"],"limitations":["Manual review required — no automated validation of prompt quality, safety, or effectiveness before merge","No contributor reputation system — all contributors are treated equally regardless of past contributions","Limited guidance on prompt quality — CONTRIBUTING.md specifies format but not content quality standards","No automated conflict resolution — duplicate or near-duplicate prompts may be submitted without detection"],"requires":["GitHub account with repository fork/PR permissions","Understanding of markdown format and YAML frontmatter","Familiarity with CONTRIBUTING.md and SECURITY.md guidelines","Git knowledge for submitting pull requests"],"input_types":["new prompt markdown files with metadata","pull request descriptions explaining contributions","security vulnerability reports"],"output_types":["merged prompt files in appropriate directory","updated TOC files (auto-generated)","contributor acknowledgment in repository"],"categories":["automation-workflow","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-louisshark--chatgpt_system_prompt__cap_7","uri":"capability://search.retrieval.prompt.file.enumeration.and.lookup","name":"prompt-file-enumeration-and-lookup","description":"Implements find_gptfile() function in idxtool.py that enables both ID-based and URL-based lookup of prompt files across the repository. The function supports searching by GPT ID (embedded in filenames like tveXvXU5g_QuantFinance.md) or by OpenAI GPT URL, returning the corresponding markdown file path and enabling programmatic access to prompt content without requiring a database backend.","intents":["I have a GPT ID or URL and need to find the corresponding system prompt in the repository","I want to programmatically access prompt files by ID for integration with other tools or scripts","I need to verify that a specific GPT's prompt is documented in the repository"],"best_for":["developers building tools that integrate with the prompt repository","researchers programmatically analyzing prompt collections","automation scripts that need to fetch specific prompts by ID","API wrappers that map GPT IDs to local prompt files"],"limitations":["Requires exact ID or URL match — no fuzzy matching or partial search capability","File-system based lookup — scales linearly with repository size, no indexing for O(1) lookup","No caching — repeated lookups require re-scanning the file system","Brittle to naming convention changes — depends on consistent filename format (ID_Name.md)"],"requires":["Python 3.6+ with file system access","Knowledge of GPT ID format or OpenAI GPT URL structure","idxtool.py script available in repository"],"input_types":["GPT ID string (e.g., 'tveXvXU5g')","OpenAI GPT URL (e.g., 'https://chatgpt.com/g/g-tveXvXU5g')"],"output_types":["file path to corresponding markdown prompt file","boolean indicating whether GPT is found in repository"],"categories":["search-retrieval","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-louisshark--chatgpt_system_prompt__cap_8","uri":"capability://memory.knowledge.official.ai.product.prompt.documentation","name":"official-ai-product-prompt-documentation","description":"Documents system prompts and configuration from official AI products including OpenAI ChatGPT, Anthropic Claude (with Code System and agent loop documentation), xAI Grok (versions 2, 3, and 3 with deep search), and Google AI tools. The /prompts/official-product/ directory includes not just system prompts but also capability descriptions, tool configurations (tools.json), and agent loop documentation, providing comprehensive insight into how commercial AI products are structured.","intents":["I want to understand how OpenAI structures ChatGPT's system prompt and instruction hierarchy","I need to study Anthropic's Claude Code System to understand code generation best practices","I want to compare Grok's jailbreak-resistant prompts across versions 2 and 3"],"best_for":["prompt engineers studying commercial AI product design","teams building AI products who want to learn from vendor approaches","researchers analyzing how different vendors implement instruction-following","developers integrating multiple AI vendors who need to understand their differences"],"limitations":["Prompts may be outdated — vendors update system prompts frequently, and repository updates are manual","Incomplete product coverage — not all vendor products or versions are documented","No official endorsement — prompts are reverse-engineered or leaked, not officially published by vendors","No configuration context — prompts are documented in isolation without information about model versions, temperature settings, or other hyperparameters"],"requires":["GitHub repository access to /prompts/official-product/ directory","Understanding of different vendor product lines and versions","Ability to interpret system prompts and configuration files"],"input_types":["official system prompts from ChatGPT, Claude, Grok, Google AI","capability descriptions and tool configurations","agent loop documentation"],"output_types":["markdown files containing system prompts","JSON configuration files (tools.json)","capability and agent loop documentation"],"categories":["memory-knowledge","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-louisshark--chatgpt_system_prompt__cap_9","uri":"capability://memory.knowledge.open.source.ai.project.prompt.collection","name":"open-source-ai-project-prompt-collection","description":"Catalogs system prompts and prompt engineering patterns from open-source AI projects (20+ entries in /prompts/opensource-prj/) including projects like II Agent, GAIA systems, and others. This collection documents how open-source developers structure prompts for specialized tasks, enabling knowledge sharing and pattern reuse across the open-source AI community.","intents":["I want to study how open-source AI projects structure their system prompts and agent loops","I need to find examples of prompt engineering patterns used in production open-source systems","I want to contribute my open-source AI project's prompts to a community collection"],"best_for":["open-source AI developers learning from peer implementations","researchers studying prompt engineering patterns in production systems","teams building open-source AI projects who want to share their approaches","developers evaluating open-source alternatives to commercial AI products"],"limitations":["Limited coverage — only 20+ open-source projects documented compared to 1,100+ custom GPTs","Inconsistent documentation quality — open-source projects may have varying levels of prompt documentation","No maintenance guarantees — open-source projects may become inactive, making their prompts outdated","No performance metrics — no information about how well these prompts perform in production"],"requires":["GitHub repository access to /prompts/opensource-prj/ directory","Understanding of open-source AI project landscape","Ability to read and interpret system prompts"],"input_types":["system prompts from open-source AI projects","agent loop and capability documentation","prompt engineering patterns and best practices"],"output_types":["markdown files containing project prompts","documentation of prompt patterns and techniques","links to original open-source projects"],"categories":["memory-knowledge","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":33,"verified":false,"data_access_risk":"high","permissions":["GitHub repository with Actions enabled","Python 3.6+ for idxtool.py and gptparser.py scripts",".github/workflows/build-toc.yaml configuration file","Standardized markdown format for prompt files with parseable metadata headers","Python 3.6+ with standard library (no external dependencies mentioned)","Markdown files with YAML frontmatter or standardized header format","Consistent field naming conventions across all prompt files","GitHub repository access to /prompts/gpts/ directory","Understanding of custom GPT architecture and system prompt design","Ability to browse and compare multiple prompt examples"],"failure_modes":["Requires GitHub Actions CI/CD integration — not portable to other version control systems without modification","TOC generation latency increases linearly with repository size (1,100+ GPTs may take 10-30 seconds per rebuild)","No incremental indexing — rebuilds entire TOC on every trigger rather than delta updates","Markdown-only output format — cannot generate alternative documentation formats (HTML, JSON indices)","Markdown-only parsing — cannot extract metadata from JSON, YAML, or other structured formats without additional parsers","Fragile to formatting variations — prompts with non-standard YAML frontmatter or missing headers may fail to parse completely","No semantic understanding of prompt content — extracts only syntactic metadata fields, not capability descriptions or use-case inference","Single-pass parsing — no error recovery or fallback strategies for malformed metadata","No quality filtering — all 1,100+ GPTs are included regardless of effectiveness or adoption","No usage metrics — cannot identify which patterns are most successful or widely used","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.35145679786952877,"quality":0.32,"ecosystem":0.48999999999999994,"match_graph":0.25,"freshness":0.75,"weights":{"adoption":0.15,"quality":0.25,"ecosystem":0.1,"match_graph":0.45,"freshness":0.05}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:22.061Z","last_scraped_at":"2026-05-03T13:59:50.673Z","last_commit":"2026-04-23T13:11:06Z"},"community":{"stars":10550,"forks":1469,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=louisshark--chatgpt_system_prompt","compare_url":"https://unfragile.ai/compare?artifact=louisshark--chatgpt_system_prompt"}},"signature":"YoL7lP5QgJELMgkhCu1uamr9M8X8FEXGu6k2Svdwkv3myrAJGOrLwtdhYu/feGoaBxUaqEBHPNOTP3fyXwYRCw==","signedAt":"2026-06-21T09:21:34.470Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/louisshark--chatgpt_system_prompt","artifact":"https://unfragile.ai/louisshark--chatgpt_system_prompt","verify":"https://unfragile.ai/api/v1/verify?slug=louisshark--chatgpt_system_prompt","publicKey":"https://unfragile.ai/api/v1/trust-passport-public-key","spec":"https://unfragile.ai/trust","schema":"https://unfragile.ai/schema.json","docs":"https://unfragile.ai/docs"}}