{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"hn-47466513","slug":"ai-sdlc-scaffold-repo-template-for-ai-assisted-sof","name":"AI SDLC Scaffold, repo template for AI-assisted software development","type":"template","url":"https://github.com/pangon/ai-sdlc-scaffold/","page_url":"https://unfragile.ai/ai-sdlc-scaffold-repo-template-for-ai-assisted-sof","categories":["app-builders"],"tags":["hackernews","show-hn"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"hn-47466513__cap_0","uri":"capability://code.generation.editing.ai.assisted.project.scaffolding.with.llm.driven.template.generation","name":"ai-assisted project scaffolding with llm-driven template generation","description":"Generates project structure, configuration files, and boilerplate code by accepting natural language project descriptions and converting them into a complete repository layout. Uses prompt engineering to guide LLMs through multi-step generation of directory hierarchies, dependency manifests, and starter code, with support for multiple tech stacks and frameworks through template composition patterns.","intents":["I want to bootstrap a new project without manually creating folder structures and config files","I need to generate initial boilerplate code that follows best practices for my chosen tech stack","I want to create a reproducible project template that can be version-controlled and reused across teams"],"best_for":["solo developers and small teams building new projects frequently","organizations standardizing on internal project templates","rapid prototyping teams that need to minimize setup time"],"limitations":["LLM-generated scaffolds may not follow all organizational conventions without explicit prompt engineering","No built-in validation that generated code compiles or passes linting without post-generation CI/CD","Template customization requires manual prompt iteration; no visual template builder","Dependency resolution relies on LLM knowledge cutoff; may suggest outdated or incompatible package versions"],"requires":["Access to an LLM API (OpenAI, Anthropic, or compatible provider)","Git for version control and template storage","Node.js 16+ or Python 3.8+ depending on implementation language","Basic understanding of project structure conventions for your tech stack"],"input_types":["natural language project description","tech stack specification (e.g., 'React + TypeScript + Tailwind')","optional: existing project templates or configuration files"],"output_types":["directory structure (file tree)","configuration files (package.json, tsconfig.json, .env.example, etc.)","starter code files (index.ts, App.tsx, main.py, etc.)","README with setup instructions"],"categories":["code-generation-editing","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"hn-47466513__cap_1","uri":"capability://automation.workflow.ai.guided.development.workflow.orchestration.with.prompt.templates","name":"ai-guided development workflow orchestration with prompt templates","description":"Provides a structured framework for integrating LLM-assisted development into the SDLC by defining prompt templates, execution patterns, and integration points for common development tasks (code review, testing, documentation). Uses a template-based approach where developers define workflows as configuration files that route code through LLM pipelines with context injection and output validation.","intents":["I want to standardize how my team uses AI tools across the development lifecycle","I need to inject project context (codebase, conventions, requirements) into every LLM interaction","I want to create reusable AI-assisted workflows that can be version-controlled and audited"],"best_for":["engineering teams adopting AI-assisted development at scale","organizations needing audit trails and reproducibility for AI-generated code","teams with established coding standards wanting to enforce them via AI"],"limitations":["Requires upfront investment in defining and testing prompt templates for your workflows","Output quality depends heavily on prompt engineering; poor templates produce poor results","No built-in feedback loop; developers must manually validate and iterate on LLM outputs","Context injection is manual; no automatic codebase indexing or semantic understanding"],"requires":["Git repository with accessible codebase","LLM API credentials (OpenAI, Anthropic, or self-hosted)","YAML or JSON configuration file support","CLI or IDE integration for workflow execution"],"input_types":["code files or snippets","prompt template definitions (YAML/JSON)","project context (README, architecture docs, style guides)","task specifications (e.g., 'review this PR', 'write tests for this function')"],"output_types":["annotated code with AI suggestions","test cases and test code","documentation and comments","code review feedback with line-by-line comments"],"categories":["automation-workflow","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"hn-47466513__cap_10","uri":"capability://automation.workflow.error.handling.and.fallback.strategies.with.graceful.degradation","name":"error handling and fallback strategies with graceful degradation","description":"Implements error handling patterns for LLM failures (rate limits, timeouts, invalid responses) with configurable fallback strategies (retry with backoff, use alternative provider, use cached response, manual intervention). Uses a resilience pattern where each workflow step has defined failure modes and recovery strategies, ensuring workflows degrade gracefully rather than failing completely.","intents":["I want workflows to handle LLM API failures without stopping the entire process","I need to implement rate limiting and retry logic without manual intervention","I want to cache LLM responses to reduce costs and latency for repeated requests"],"best_for":["production systems requiring high availability","teams with cost constraints needing to minimize API calls","organizations with strict SLAs requiring graceful degradation"],"limitations":["Fallback strategies are manual; no automatic selection based on failure type","Caching requires storage infrastructure; no built-in persistence","Retry logic adds latency; exponential backoff can cause long delays","Manual intervention fallbacks require human availability"],"requires":["Error handling configuration (retry policies, fallback strategies)","Optional: cache storage (Redis, database, file system)","Optional: monitoring and alerting for manual intervention","LLM API with rate limit information"],"input_types":["error configuration (retry count, backoff strategy, fallback rules)","workflow definitions with error handlers","optional: cache configuration"],"output_types":["workflow results (from primary or fallback path)","error logs with recovery information","metrics (success rate, fallback usage, cache hit rate)"],"categories":["automation-workflow","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"hn-47466513__cap_11","uri":"capability://safety.moderation.output.validation.and.quality.gates.with.structured.schema.enforcement","name":"output validation and quality gates with structured schema enforcement","description":"Validates LLM outputs against defined schemas (JSON, code syntax, format requirements) and quality criteria (length, complexity, coverage) before accepting them into workflows. Uses a validation layer where outputs are checked against schemas and rules, with failures triggering re-generation, manual review, or fallback strategies. Supports structured outputs (JSON, code) with schema validation and unstructured outputs (text) with regex or semantic validation.","intents":["I want to ensure generated code is syntactically valid before using it","I need to validate that generated outputs meet quality criteria (e.g., test coverage, documentation completeness)","I want to catch invalid outputs early and trigger re-generation or manual review"],"best_for":["production systems requiring high output quality","teams with strict quality standards and compliance requirements","workflows where invalid outputs cause downstream failures"],"limitations":["Schema validation is strict; may reject valid outputs that don't match schema exactly","Quality criteria are manual; no automatic scoring of subjective qualities","Re-generation on failure increases latency and API costs","Semantic validation (e.g., 'is this code idiomatic?') is difficult without human review"],"requires":["Output schema definitions (JSON Schema, regex patterns, etc.)","Quality criteria definitions (metrics, thresholds)","Optional: code parser/linter for syntax validation","Optional: human review process for failed validations"],"input_types":["LLM outputs (code, JSON, text)","schema definitions (JSON Schema, regex, custom rules)","quality criteria (metrics, thresholds)"],"output_types":["validation results (pass/fail with details)","corrected outputs (if re-generation is triggered)","validation reports with metrics"],"categories":["safety-moderation","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"hn-47466513__cap_12","uri":"capability://automation.workflow.team.collaboration.features.with.shared.prompt.libraries.and.audit.trails","name":"team collaboration features with shared prompt libraries and audit trails","description":"Enables team collaboration on AI workflows by providing shared prompt libraries, version control for prompts and configurations, and audit trails showing who made what changes and when. Uses a centralized repository pattern where prompts, workflows, and configurations are stored with metadata (author, timestamp, change description), enabling teams to collaborate on AI development similar to code collaboration.","intents":["I want my team to share and reuse prompts without duplicating effort","I need to track who made changes to prompts and why for compliance or debugging","I want to prevent conflicting changes to shared workflows"],"best_for":["teams of 3+ developers working on AI-assisted workflows","organizations with compliance requirements needing audit trails","teams wanting to build institutional knowledge of effective prompts"],"limitations":["Shared libraries require discipline to maintain; outdated prompts can cause confusion","Audit trails add overhead; no automatic cleanup of old versions","Conflict resolution for concurrent prompt edits is manual","No built-in access control; requires external permission management"],"requires":["Centralized storage (Git, database, or file server)","Version control system (Git, or custom versioning)","Optional: access control system (RBAC, LDAP, etc.)","Optional: notification system for changes"],"input_types":["prompt templates and configurations","change metadata (author, timestamp, description)","optional: access control rules"],"output_types":["shared prompt library (versioned)","audit logs (who changed what when)","change notifications and diffs"],"categories":["automation-workflow","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"hn-47466513__cap_2","uri":"capability://memory.knowledge.codebase.context.injection.for.llm.interactions.with.semantic.awareness","name":"codebase context injection for llm interactions with semantic awareness","description":"Automatically extracts and injects relevant project context (architecture docs, code examples, style guides, dependency information) into LLM prompts to improve code generation quality. Uses file-based context selection patterns where developers specify which files/directories are relevant to a task, and the system prepends them to prompts with structural markers to help LLMs understand project conventions.","intents":["I want the AI to understand my project's architecture and conventions without manually explaining them each time","I need to ensure generated code follows my team's style guide and patterns","I want to reduce hallucinations by grounding LLM responses in actual project code"],"best_for":["teams with established architectural patterns and coding standards","large codebases where context is critical for code quality","organizations using multiple LLM providers and wanting consistent context handling"],"limitations":["Context window limits mean only a subset of codebase can be injected; requires manual selection or heuristics","No semantic understanding of code; uses file-based or regex-based selection rather than AST analysis","Stale context if codebase changes frequently; requires periodic re-indexing","Large context payloads increase API costs and latency proportionally"],"requires":["Git repository with readable source files","Configuration file defining context sources (paths, file patterns)","LLM with sufficient context window (4K+ tokens recommended)","Optional: documentation files (README, ARCHITECTURE.md, STYLE_GUIDE.md)"],"input_types":["project file paths or glob patterns","documentation files (markdown, text)","code examples or reference implementations","configuration files (package.json, requirements.txt, etc.)"],"output_types":["augmented prompts with injected context","context metadata (file count, token count, relevance scores)"],"categories":["memory-knowledge","code-generation-editing"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"hn-47466513__cap_3","uri":"capability://planning.reasoning.multi.step.ai.task.decomposition.with.intermediate.validation","name":"multi-step ai task decomposition with intermediate validation","description":"Breaks down complex development tasks (e.g., 'implement authentication system') into smaller LLM-solvable steps with validation gates between each step. Uses a chain-of-thought pattern where each step produces intermediate artifacts (design docs, code sketches, test plans) that are validated before proceeding to the next step, reducing hallucinations and improving overall quality.","intents":["I want to tackle complex features without overwhelming the LLM with a single large prompt","I need to validate AI-generated designs before committing to implementation","I want to create audit trails showing how AI arrived at final code"],"best_for":["teams implementing complex features with multiple interdependent components","organizations requiring design review before implementation","developers learning from AI-generated code who want to understand reasoning"],"limitations":["Multi-step workflows increase total API calls and latency; no parallelization support","Validation gates are manual; no automated correctness checking","Intermediate artifacts must be manually reviewed; no automatic quality scoring","Task decomposition itself requires manual specification; no automatic breakdown"],"requires":["LLM API with support for multiple sequential calls","Workflow definition format (YAML, JSON, or custom DSL)","Human reviewer availability for validation gates","Version control for tracking intermediate artifacts"],"input_types":["high-level task description","acceptance criteria or requirements","optional: reference implementations or design patterns","validation rules or quality criteria"],"output_types":["design documents or architecture sketches","implementation code with comments","test cases and test code","validation reports at each step"],"categories":["planning-reasoning","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"hn-47466513__cap_4","uri":"capability://code.generation.editing.ai.assisted.code.review.with.pattern.based.feedback.generation","name":"ai-assisted code review with pattern-based feedback generation","description":"Analyzes code changes against project conventions, best practices, and custom rules by feeding diffs and context to LLMs, which generate structured feedback with specific line-by-line comments and suggestions. Uses a template-based approach where review criteria (security, performance, style, testing) are defined as prompts that guide the LLM to produce consistent, actionable feedback.","intents":["I want to automate routine code review tasks while keeping human reviewers for complex decisions","I need to enforce coding standards consistently across pull requests","I want to catch common mistakes (missing error handling, security issues) before human review"],"best_for":["teams with high PR volume and established coding standards","organizations wanting to reduce code review latency for junior developers","projects with security or compliance requirements"],"limitations":["LLM-generated feedback can be generic or miss context-specific issues","No integration with GitHub/GitLab APIs; requires manual diff extraction","Cannot detect runtime issues or performance regressions; only static analysis","False positives require manual filtering; no feedback loop to improve prompts"],"requires":["Git diff or code change data","LLM API with code understanding capabilities","Project context (style guide, architecture docs, test coverage requirements)","Optional: integration with GitHub/GitLab for automated PR comments"],"input_types":["code diffs (unified diff format)","full file context (before/after)","review criteria definitions (YAML/JSON)","project conventions and style guides"],"output_types":["structured feedback (JSON with line numbers, severity, suggestions)","formatted PR comments (markdown)","summary report with statistics (issues found, categories)"],"categories":["code-generation-editing","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"hn-47466513__cap_5","uri":"capability://code.generation.editing.test.generation.from.code.and.requirements.with.coverage.tracking","name":"test generation from code and requirements with coverage tracking","description":"Generates unit tests, integration tests, and edge case tests by analyzing code structure and requirements, then producing test code that covers specified coverage targets. Uses LLM-based test generation where prompts include the function/module to test, existing tests as examples, and coverage goals, producing executable test code in the project's test framework.","intents":["I want to generate test cases for functions without manually writing boilerplate","I need to improve test coverage quickly without sacrificing quality","I want to ensure edge cases and error conditions are tested"],"best_for":["teams with low test coverage wanting to improve quickly","developers writing tests for legacy code without existing test suites","projects with strict coverage requirements (e.g., 80%+ coverage)"],"limitations":["Generated tests may not catch all real-world edge cases; require manual review","Test quality depends on code clarity; poorly documented functions produce weak tests","No integration with coverage tools; coverage tracking is manual","Framework-specific; requires separate prompts for Jest, pytest, JUnit, etc."],"requires":["Source code files with clear function signatures","Test framework (Jest, pytest, JUnit, etc.) installed","LLM API with code understanding","Optional: existing test examples for in-context learning"],"input_types":["source code (functions, classes, modules)","requirements or docstrings","existing test examples (for pattern matching)","coverage targets (e.g., 'achieve 80% line coverage')"],"output_types":["test code in project's test framework","test execution results (pass/fail)","coverage reports (line, branch, function coverage)"],"categories":["code-generation-editing","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"hn-47466513__cap_6","uri":"capability://text.generation.language.documentation.generation.from.code.with.architecture.aware.summaries","name":"documentation generation from code with architecture-aware summaries","description":"Automatically generates README files, API documentation, and architecture guides by analyzing code structure, comments, and project metadata. Uses LLM-based documentation generation where the system extracts code structure (functions, classes, modules), existing comments, and project context, then generates human-readable documentation with examples and usage patterns.","intents":["I want to generate API documentation without manually writing docstrings for every function","I need to create architecture documentation that stays in sync with code changes","I want to generate onboarding guides for new developers joining the project"],"best_for":["teams with large codebases lacking documentation","open-source projects needing comprehensive README and API docs","organizations with high developer turnover needing good onboarding materials"],"limitations":["Generated documentation may be generic; requires manual editing for clarity","No automatic sync with code changes; documentation becomes stale","Cannot infer business logic from code alone; requires comments or docstrings","Formatting and structure depend on LLM output; inconsistent across projects"],"requires":["Source code with reasonable structure and naming conventions","LLM API with code understanding","Optional: existing documentation for style matching","Optional: project metadata (package.json, setup.py, etc.)"],"input_types":["source code files (entire modules or directories)","existing comments and docstrings","project metadata (name, version, description)","optional: architecture diagrams or design docs"],"output_types":["README.md with project overview and setup instructions","API documentation (markdown or HTML)","Architecture guides and design documents","Usage examples and tutorials"],"categories":["text-generation-language","code-generation-editing"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"hn-47466513__cap_7","uri":"capability://automation.workflow.git.integrated.workflow.automation.with.commit.level.ai.analysis","name":"git-integrated workflow automation with commit-level ai analysis","description":"Integrates AI analysis into Git workflows by analyzing commits, pull requests, and branches to generate commit messages, detect breaking changes, and suggest refactoring opportunities. Uses Git hooks and metadata to trigger LLM analysis at key points (pre-commit, pre-push, PR creation), producing structured outputs that inform development decisions.","intents":["I want to generate meaningful commit messages automatically from code changes","I need to detect breaking changes and API incompatibilities before merging","I want to identify refactoring opportunities and technical debt in pull requests"],"best_for":["teams with large codebases where commit history is important","organizations enforcing conventional commits or semantic versioning","projects with strict API compatibility requirements"],"limitations":["Commit message generation requires clear, focused commits; works poorly with large multi-file changes","Breaking change detection relies on code analysis; may miss implicit API changes","Git hook integration requires local setup; not enforced for all developers","No integration with GitHub/GitLab workflows; requires custom CI/CD setup"],"requires":["Git repository with accessible commit history","Git hooks support (pre-commit, pre-push, etc.)","LLM API with code understanding","Optional: GitHub/GitLab API for PR integration"],"input_types":["Git diffs (staged changes)","commit history (previous commits for context)","branch metadata (base branch, target branch)","optional: API schema or version information"],"output_types":["generated commit messages (conventional format)","breaking change reports","refactoring suggestions with code examples","changelog entries"],"categories":["automation-workflow","code-generation-editing"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"hn-47466513__cap_8","uri":"capability://tool.use.integration.configuration.driven.llm.provider.abstraction.with.multi.provider.support","name":"configuration-driven llm provider abstraction with multi-provider support","description":"Abstracts LLM provider differences (OpenAI, Anthropic, local models) behind a unified interface, allowing workflows to switch providers via configuration without code changes. Uses a provider adapter pattern where each LLM provider implements a standard interface (prompt submission, response parsing, token counting), and a configuration layer routes requests to the appropriate provider based on task requirements.","intents":["I want to use different LLM providers for different tasks without rewriting workflows","I need to switch providers (e.g., from OpenAI to Anthropic) without changing code","I want to run workflows on local models for privacy or cost reasons"],"best_for":["organizations evaluating multiple LLM providers","teams with cost or privacy constraints requiring provider flexibility","projects needing to support both cloud and self-hosted models"],"limitations":["API differences between providers require normalization; some features may not be available across all providers","Token counting and rate limiting differ by provider; no unified cost tracking","Model capabilities vary (context window, reasoning, code understanding); workflows may behave differently across providers","Fallback logic is manual; no automatic provider failover"],"requires":["Configuration file (YAML/JSON) defining provider settings","API credentials for each provider (OpenAI key, Anthropic key, etc.)","Optional: local LLM setup (Ollama, LM Studio, etc.) for self-hosted models","Adapter implementations for each provider"],"input_types":["provider configuration (model name, API endpoint, credentials)","prompts and messages (standard format)","task metadata (required capabilities, cost constraints)"],"output_types":["LLM responses (normalized format)","usage statistics (tokens, cost, latency)","provider metadata (model version, capabilities)"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"hn-47466513__cap_9","uri":"capability://automation.workflow.prompt.versioning.and.experimentation.with.a.b.testing.support","name":"prompt versioning and experimentation with a/b testing support","description":"Enables version control and experimentation for prompts by storing prompt templates with metadata (version, author, performance metrics) and supporting A/B testing workflows where different prompt versions are tested against the same input. Uses a prompt registry pattern where prompts are stored as versioned artifacts with associated metrics, enabling data-driven prompt optimization.","intents":["I want to track how prompt changes affect code generation quality","I need to compare different prompt strategies (e.g., few-shot vs. chain-of-thought) systematically","I want to version prompts alongside code and track which version produced which output"],"best_for":["teams optimizing AI workflows through experimentation","organizations with strict quality requirements needing data-driven decisions","research teams studying prompt engineering techniques"],"limitations":["A/B testing requires manual evaluation of outputs; no automatic quality scoring","Metrics collection is manual; no built-in instrumentation","Statistical significance requires large sample sizes; impractical for small teams","Prompt versioning adds complexity; requires discipline to maintain"],"requires":["Prompt storage system (database, file system, or Git)","Metrics collection and analysis tools (optional)","LLM API for running experiments","Test dataset or evaluation criteria"],"input_types":["prompt templates (text with variables)","prompt metadata (version, author, description)","test inputs (code, requirements, etc.)","evaluation criteria or quality metrics"],"output_types":["versioned prompt artifacts","experiment results (outputs from each prompt version)","performance metrics (quality scores, latency, cost)","comparison reports (which prompt performed best)"],"categories":["automation-workflow","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":37,"verified":false,"data_access_risk":"high","permissions":["Access to an LLM API (OpenAI, Anthropic, or compatible provider)","Git for version control and template storage","Node.js 16+ or Python 3.8+ depending on implementation language","Basic understanding of project structure conventions for your tech stack","Git repository with accessible codebase","LLM API credentials (OpenAI, Anthropic, or self-hosted)","YAML or JSON configuration file support","CLI or IDE integration for workflow execution","Error handling configuration (retry policies, fallback strategies)","Optional: cache storage (Redis, database, file system)"],"failure_modes":["LLM-generated scaffolds may not follow all organizational conventions without explicit prompt engineering","No built-in validation that generated code compiles or passes linting without post-generation CI/CD","Template customization requires manual prompt iteration; no visual template builder","Dependency resolution relies on LLM knowledge cutoff; may suggest outdated or incompatible package versions","Requires upfront investment in defining and testing prompt templates for your workflows","Output quality depends heavily on prompt engineering; poor templates produce poor results","No built-in feedback loop; developers must manually validate and iterate on LLM outputs","Context injection is manual; no automatic codebase indexing or semantic understanding","Fallback strategies are manual; no automatic selection based on failure type","Caching requires storage infrastructure; no built-in persistence","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.46,"quality":0.35,"ecosystem":0.46,"match_graph":0.25,"freshness":0.6,"weights":{"adoption":0.25,"quality":0.25,"ecosystem":0.1,"match_graph":0.35,"freshness":0.05}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-06-17T09:51:04.692Z","last_scraped_at":"2026-05-04T08:09:54.664Z","last_commit":null},"community":{"stars":null,"forks":null,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=ai-sdlc-scaffold-repo-template-for-ai-assisted-sof","compare_url":"https://unfragile.ai/compare?artifact=ai-sdlc-scaffold-repo-template-for-ai-assisted-sof"}},"signature":"Er+tIj0U9LRGToEr0ZVaAeQKJERo+JN3zCt2Odx4HQ+aI/Z4V0ESnfoLaMHqelaGZMjcLLUp6ZhRNdQcz+oNBw==","signedAt":"2026-06-20T21:39:00.296Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/ai-sdlc-scaffold-repo-template-for-ai-assisted-sof","artifact":"https://unfragile.ai/ai-sdlc-scaffold-repo-template-for-ai-assisted-sof","verify":"https://unfragile.ai/api/v1/verify?slug=ai-sdlc-scaffold-repo-template-for-ai-assisted-sof","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"}}