{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"github-getagentseal--codeburn","slug":"getagentseal--codeburn","name":"codeburn","type":"cli","url":"https://github.com/getagentseal/codeburn","page_url":"https://unfragile.ai/getagentseal--codeburn","categories":["observability","app-builders"],"tags":["ai-coding","claude-code","cli","codex","cost-tracking","cursor-ide","developer-tools","observability","terminal-ui","token-usage"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"github-getagentseal--codeburn__cap_0","uri":"capability://data.processing.analysis.multi.provider.session.log.discovery.and.parsing","name":"multi-provider session log discovery and parsing","description":"Automatically locates and parses session logs from Claude Code, Cursor, GitHub Copilot, Codex, and other AI coding tools by scanning platform-specific directories (~/.claude, ~/.config, etc.). Implements a provider plugin system with standardized parsers that convert heterogeneous log formats into a unified ParsedTurn and Session object model, enabling downstream analysis across multiple tools without manual configuration.","intents":["I want to track token usage across all my AI coding tools in one place without manually collecting logs","I need to parse session data from Claude Code and Cursor simultaneously to compare their efficiency","I want to add support for a new AI coding provider without rewriting the entire parsing pipeline"],"best_for":["developers using multiple AI coding assistants (Claude Code + Cursor + Copilot)","teams auditing AI tool usage across different IDEs","tool builders extending CodeBurn with custom provider support"],"limitations":["Requires local access to session log directories — cannot parse logs from cloud-only tools without local caching","Parser accuracy depends on provider log format stability — breaking changes in IDE session formats require parser updates","No real-time streaming of logs — only processes completed session files on disk"],"requires":["Node.js 22+","macOS or Linux (for POSIX path resolution to ~/.claude, ~/.config, etc.)","Write permissions to cache directory for parsed session metadata"],"input_types":["JSON session logs from Claude Code","SQLite databases from Cursor IDE","Log files from GitHub Copilot","Structured session metadata from other providers"],"output_types":["ParsedTurn objects (standardized turn representation)","Session objects (grouped turns with metadata)","Aggregated project-level session collections"],"categories":["data-processing-analysis","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-getagentseal--codeburn__cap_1","uri":"capability://data.processing.analysis.turn.classification.and.task.categorization.engine","name":"turn classification and task categorization engine","description":"Analyzes parsed session turns and classifies them into TaskCategory buckets (coding, testing, terminal usage, debugging, etc.) using heuristic rules based on turn content, tool invocations, and file types. Implements a classifyTurn function that examines API calls, file modifications, and context patterns to assign semantic meaning to raw token consumption, enabling cost breakdown by activity type rather than just by model.","intents":["I want to see how much I'm spending on testing vs actual coding tasks","I need to identify which task categories are burning the most tokens in my workflow","I want to understand if my AI assistant is being used efficiently for its intended purpose"],"best_for":["developers optimizing their AI coding workflow by task type","engineering managers analyzing team AI tool usage patterns","researchers studying how developers interact with AI coding assistants"],"limitations":["Classification is heuristic-based and may misclassify edge cases (e.g., debugging-heavy coding tasks)","Requires sufficient turn context to classify accurately — minimal turns with sparse metadata may default to generic categories","No machine learning model — classification rules are hand-crafted and may not capture domain-specific task patterns"],"requires":["Parsed session data with turn content and API call metadata","File type information from the codebase being edited","Tool invocation records (e.g., MCP tool calls, terminal commands)"],"input_types":["ParsedTurn objects with content and metadata","File paths and extensions from edited files","Tool invocation logs and MCP function calls"],"output_types":["TaskCategory enum (coding, testing, terminal, debugging, refactoring, etc.)","Classified turn objects with category metadata","Cost breakdown by task category"],"categories":["data-processing-analysis","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-getagentseal--codeburn__cap_10","uri":"capability://data.processing.analysis.status.reporting.and.session.metadata.inspection","name":"status reporting and session metadata inspection","description":"Provides CLI commands (codeburn status, codeburn report) that generate detailed reports on session discovery status, parsing errors, and data quality metrics. Implements metadata inspection capabilities that allow developers to examine individual session files, view parsing errors, and understand data completeness. Generates status summaries showing how many sessions were discovered, parsed successfully, and skipped due to errors.","intents":["I want to verify that CodeBurn is correctly discovering all my AI coding sessions","I need to debug why some sessions aren't being parsed or included in my cost calculations","I want to understand the quality and completeness of my session data"],"best_for":["developers troubleshooting CodeBurn setup or data discovery issues","teams auditing data quality before relying on CodeBurn for cost reporting","developers integrating CodeBurn into monitoring systems that need status checks"],"limitations":["Status reports are point-in-time snapshots — do not track changes over time without manual comparison","Parsing error details may be verbose and difficult to interpret for non-technical users","No automatic remediation — errors are reported but require manual investigation"],"requires":["Session discovery and parsing pipeline","Error logging from parsers","Metadata about discovered sessions"],"input_types":["Session discovery results","Parsing logs and error messages","Session metadata"],"output_types":["Status summary (sessions discovered, parsed, skipped)","Detailed error reports","Data quality metrics"],"categories":["data-processing-analysis","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-getagentseal--codeburn__cap_11","uri":"capability://tool.use.integration.provider.plugin.system.with.extensible.architecture","name":"provider plugin system with extensible architecture","description":"Implements a plugin-based architecture that allows new AI coding providers to be added without modifying core CodeBurn code. Each provider plugin implements standardized interfaces (discoverAllSessions, parseSessionFile) that return normalized ParsedTurn and Session objects. Plugins are loaded dynamically at runtime and can be distributed as npm packages, enabling community contributions and custom provider support.","intents":["I want to add support for a new AI coding tool that CodeBurn doesn't currently support","I need to create a custom provider plugin for my organization's internal AI tool","I want to extend CodeBurn with provider-specific optimizations without forking the project"],"best_for":["tool builders extending CodeBurn with custom provider support","organizations using proprietary AI coding tools that need CodeBurn integration","open-source contributors adding support for new AI coding platforms"],"limitations":["Plugin development requires understanding the ParsedTurn/Session data model — documentation must be comprehensive","No built-in plugin marketplace or registry — plugins must be manually installed and configured","Plugin compatibility is not guaranteed across CodeBurn versions — breaking changes to core interfaces require plugin updates"],"requires":["TypeScript or JavaScript knowledge","Understanding of the ParsedTurn and Session data models","Access to the target provider's session log format or API"],"input_types":["Provider-specific session log formats","Plugin interface definitions"],"output_types":["Normalized ParsedTurn and Session objects","Plugin metadata (name, version, provider)"],"categories":["tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-getagentseal--codeburn__cap_2","uri":"capability://data.processing.analysis.accurate.cost.calculation.with.litellm.pricing.integration","name":"accurate cost calculation with litellm pricing integration","description":"Calculates USD costs for each turn by multiplying token counts (input + output) by model-specific pricing rates sourced from LiteLLM's pricing database, which covers 100+ models across OpenAI, Anthropic, and other providers. Implements a calculateCost function that handles variable pricing tiers, currency conversion, and subscription plan adjustments (e.g., Claude Pro discounts), ensuring accurate financial visibility without requiring API calls to pricing services.","intents":["I want to know the exact USD cost of each AI coding session without guessing","I need to track whether my Claude Pro subscription is actually saving me money","I want to compare the cost-effectiveness of different models (GPT-4 vs Claude 3.5 Sonnet) based on my actual usage"],"best_for":["individual developers managing personal AI tool budgets","engineering teams tracking AI tool spend against departmental budgets","companies evaluating ROI of AI coding tool subscriptions"],"limitations":["Pricing data is static and updated only when LiteLLM releases new pricing — real-time pricing changes are not reflected until CodeBurn updates","Does not account for volume discounts or enterprise pricing agreements — only supports standard public pricing","Currency conversion uses fixed rates at data ingestion time — does not reflect real-time forex fluctuations"],"requires":["Token counts from parsed session turns (input_tokens, output_tokens)","Model identifier (e.g., 'claude-3-5-sonnet-20241022')","LiteLLM pricing database (bundled or fetched at runtime)","Optional: subscription plan type for discount calculation"],"input_types":["Parsed turn objects with token counts","Model identifiers from session metadata","Subscription plan information (free, Claude Pro, etc.)"],"output_types":["USD cost per turn (float)","Aggregated cost by model, project, or time period","Cost comparison matrices between models"],"categories":["data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-getagentseal--codeburn__cap_3","uri":"capability://automation.workflow.interactive.terminal.ui.dashboard.with.real.time.metrics","name":"interactive terminal ui dashboard with real-time metrics","description":"Renders a terminal-based interactive dashboard (TUI) using a framework like Ink or Blessed that displays aggregated token usage, costs, and efficiency metrics across multiple time periods (Today, 7 Days, 30 Days, All Time). Implements keyboard-driven navigation, filtering by project/model/task category, and drill-down capabilities that allow developers to explore cost patterns without leaving the terminal. Updates metrics in real-time as new session data is discovered.","intents":["I want to quickly check my AI token burn rate without opening a web browser","I need to drill down from monthly costs to individual turns to find where tokens are being wasted","I want to compare my token usage across projects and models in an interactive interface"],"best_for":["developers working in terminal-first environments (SSH, remote development)","teams using CodeBurn in CI/CD pipelines for cost monitoring","developers who prefer keyboard-driven interfaces over GUI applications"],"limitations":["Terminal rendering is limited by screen size — complex datasets with many projects/models may be difficult to navigate","No mouse support in most terminal environments — navigation is keyboard-only","Performance degrades with very large datasets (1000+ turns) — rendering may become slow without pagination"],"requires":["Terminal emulator with ANSI color support (most modern terminals)","Node.js 22+ runtime","Aggregated session data from the day aggregation pipeline"],"input_types":["Aggregated cost data by project, model, task category","Time-period summaries (Today, 7 Days, etc.)","Individual turn records for drill-down"],"output_types":["Rendered terminal UI with tables, charts, and metrics","User navigation state (selected filters, drill-down level)","Export triggers (CSV, JSON)"],"categories":["automation-workflow","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-getagentseal--codeburn__cap_4","uri":"capability://data.processing.analysis.daily.aggregation.and.time.period.bucketing.with.caching","name":"daily aggregation and time-period bucketing with caching","description":"Aggregates parsed session turns into daily buckets and higher-level time periods (7 Days, 30 Days, All Time) using an aggregateProjectsIntoDays function that groups by date, project, and model. Implements a caching layer that stores aggregated results to avoid recomputing statistics on every dashboard load, with cache invalidation triggered by new session data discovery. Supports efficient querying of cost trends across arbitrary time windows.","intents":["I want to see my daily token burn rate without recalculating from raw turns every time","I need to compare my spending across weeks and months to identify trends","I want to cache aggregated data so the dashboard loads quickly even with thousands of turns"],"best_for":["developers with large session histories (1000+ turns) who need fast dashboard performance","teams running CodeBurn in monitoring systems that need sub-second metric queries","developers analyzing long-term AI tool usage trends"],"limitations":["Cache invalidation is event-driven (new session discovery) — manual cache clearing may be required if session files are modified externally","Aggregation is immutable per day — retroactive corrections to historical costs require cache rebuild","Time-period bucketing uses fixed boundaries (calendar days) — custom time windows require on-the-fly aggregation without cache benefit"],"requires":["Parsed session turns with timestamps and costs","Persistent cache storage (file system or in-memory)","Date/time library for bucketing logic"],"input_types":["ParsedTurn objects with timestamps and costs","Project and model metadata"],"output_types":["DayAggregation objects (daily cost summaries by project/model)","PeriodData objects (7-day, 30-day, all-time summaries)","Cached aggregation results"],"categories":["data-processing-analysis","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-getagentseal--codeburn__cap_5","uri":"capability://planning.reasoning.token.burn.pattern.detection.and.optimization.recommendations","name":"token burn pattern detection and optimization recommendations","description":"Scans session history to identify inefficient token usage patterns such as redundant file reads, bloated context windows, unused MCP tool invocations, and low one-shot success rates. Implements an optimization engine (codeburn optimize) that analyzes turn sequences, detects repeated operations on the same files, and generates actionable recommendations to reduce token waste. Uses heuristic rules and statistical analysis to flag anomalies in token consumption.","intents":["I want to find where I'm wasting tokens in my AI coding workflow","I need to understand why my one-shot success rate is low and how to improve it","I want recommendations on how to reduce my AI tool costs without sacrificing productivity"],"best_for":["developers with high token burn rates looking to optimize their workflow","engineering teams analyzing inefficiencies in AI tool usage","researchers studying token waste patterns in AI-assisted development"],"limitations":["Pattern detection is heuristic-based and may produce false positives (e.g., legitimate repeated file reads)","Recommendations are generic and may not account for domain-specific constraints (e.g., large codebases requiring large context)","Requires sufficient historical data (100+ turns) to detect meaningful patterns — small session histories may not yield actionable insights"],"requires":["Parsed session turns with file access logs and tool invocations","One-shot success rate metrics per turn","Historical session data spanning multiple days"],"input_types":["ParsedTurn objects with file access patterns","MCP tool invocation logs","Turn success/failure indicators"],"output_types":["Detected waste patterns (redundant reads, bloated context, etc.)","Optimization recommendations with estimated token savings","Anomaly flags for unusual token consumption"],"categories":["planning-reasoning","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-getagentseal--codeburn__cap_6","uri":"capability://data.processing.analysis.model.comparison.and.cost.effectiveness.analysis","name":"model comparison and cost-effectiveness analysis","description":"Implements a model comparison engine (codeburn compare) that analyzes token usage and costs across different AI models used in the same session history, calculating metrics like cost-per-successful-turn, average context size, and one-shot success rate per model. Generates comparison matrices and visualizations that help developers understand which models are most cost-effective for their specific workflow, accounting for both raw cost and task completion efficiency.","intents":["I want to know if Claude 3.5 Sonnet is more cost-effective than GPT-4 for my coding tasks","I need to compare the one-shot success rate across different models to understand quality vs cost tradeoffs","I want to see which model gives me the best cost-per-completed-task ratio"],"best_for":["developers evaluating which AI model to standardize on","teams making procurement decisions between Claude, OpenAI, and other providers","researchers studying cost-effectiveness of different AI coding models"],"limitations":["Comparison is only valid if models were used on similar tasks — comparing Claude on refactoring vs GPT-4 on debugging is not meaningful","Requires sufficient usage of each model (50+ turns) to produce statistically valid comparisons","Does not account for qualitative factors like code quality, security, or maintainability — only quantifies cost and speed metrics"],"requires":["Session history with multiple models used","Cost data per turn (from cost calculation capability)","One-shot success rate metrics per turn","Task category classification for fair comparison"],"input_types":["Aggregated cost and metric data by model","One-shot success rates per model","Task category distribution per model"],"output_types":["Comparison matrices (cost, success rate, context size by model)","Cost-effectiveness rankings","Visualization data for charts and graphs"],"categories":["data-processing-analysis","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-getagentseal--codeburn__cap_7","uri":"capability://data.processing.analysis.csv.and.structured.data.export.with.custom.filtering","name":"csv and structured data export with custom filtering","description":"Exports aggregated cost data and detailed turn records to CSV and JSON formats with support for custom filtering by project, model, task category, date range, and other dimensions. Implements an exportCsv function that generates spreadsheet-compatible output suitable for further analysis in Excel, Google Sheets, or data analysis tools. Supports both summary-level exports (daily aggregates) and detailed exports (individual turns with full metadata).","intents":["I want to export my token usage data to Excel for reporting to my manager","I need to analyze my AI tool costs in a spreadsheet with custom formulas","I want to share detailed cost breakdowns with my team in a portable format"],"best_for":["developers creating reports for management or finance teams","teams performing custom analysis of AI tool costs in spreadsheet tools","organizations integrating CodeBurn data into existing BI/analytics systems"],"limitations":["CSV export flattens hierarchical data (e.g., turns within projects) — complex relationships may be lost","Large exports (10,000+ rows) may be slow to generate and difficult to open in spreadsheet applications","No built-in scheduling for automated exports — exports are manual or require external orchestration"],"requires":["Aggregated or detailed cost data","CSV/JSON serialization library","File system write permissions"],"input_types":["Aggregated cost data by project/model/category","Individual turn records with full metadata","Filter criteria (date range, project, model, etc.)"],"output_types":["CSV files with headers and data rows","JSON files with structured hierarchies","Exported file paths"],"categories":["data-processing-analysis","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-getagentseal--codeburn__cap_8","uri":"capability://automation.workflow.macos.menubar.application.with.persistent.monitoring","name":"macos menubar application with persistent monitoring","description":"Provides a native macOS menubar application that displays real-time token usage and cost metrics in the system menu bar, with a dropdown menu showing today's spend, weekly trends, and quick access to the full dashboard. Implements app lifecycle management, state persistence, and background monitoring that updates metrics as new session data is discovered. Uses native macOS UI frameworks for seamless integration with the operating system.","intents":["I want to see my AI token burn rate at a glance from the macOS menu bar","I need quick access to cost metrics without opening a terminal or browser","I want to receive notifications when my daily token spend exceeds a threshold"],"best_for":["macOS developers who want always-visible cost monitoring","teams using CodeBurn on macOS workstations for real-time observability","developers who prefer native applications over terminal or web interfaces"],"limitations":["macOS-only — no support for Linux or Windows","Menubar space is limited — only high-level metrics can be displayed without opening the full app","Background monitoring adds system resource overhead — may impact battery life on laptops"],"requires":["macOS 10.13+ (or version specified in package.json)","Electron or native macOS framework for app development","Session data from local AI coding tools"],"input_types":["Aggregated cost metrics from the analysis pipeline","Real-time session data updates"],"output_types":["Rendered menubar icon with current metrics","Dropdown menu with summary data","Full dashboard window on demand"],"categories":["automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-getagentseal--codeburn__cap_9","uri":"capability://data.processing.analysis.subscription.plan.and.currency.configuration.management","name":"subscription plan and currency configuration management","description":"Manages subscription plan settings (free, Claude Pro, enterprise) and currency preferences (USD, EUR, GBP, etc.) that affect cost calculations and display. Implements a configuration system that stores user preferences persistently and applies subscription-specific discounts (e.g., Claude Pro 20% reduction) and currency conversion to all cost metrics. Supports multi-currency display for international teams.","intents":["I want to track my costs in EUR instead of USD since I'm in Europe","I need to apply my Claude Pro subscription discount to my cost calculations","I want to configure CodeBurn once and have my preferences persist across sessions"],"best_for":["international developers using AI tools in non-USD currencies","developers with paid subscriptions (Claude Pro, etc.) who want accurate cost tracking","teams managing CodeBurn across multiple users with different preferences"],"limitations":["Currency conversion uses fixed rates at configuration time — does not reflect real-time forex changes","Subscription discounts are hardcoded for known plans — custom enterprise discounts require manual configuration","Configuration is per-user — no team-wide settings synchronization"],"requires":["Configuration file storage (JSON, YAML, or environment variables)","Currency conversion library or static rate table","Subscription plan definitions"],"input_types":["User preference settings (currency, subscription plan)","Cost data in base currency (USD)"],"output_types":["Converted costs in user's preferred currency","Applied subscription discounts","Persisted configuration"],"categories":["data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":50,"verified":false,"data_access_risk":"high","permissions":["Node.js 22+","macOS or Linux (for POSIX path resolution to ~/.claude, ~/.config, etc.)","Write permissions to cache directory for parsed session metadata","Parsed session data with turn content and API call metadata","File type information from the codebase being edited","Tool invocation records (e.g., MCP tool calls, terminal commands)","Session discovery and parsing pipeline","Error logging from parsers","Metadata about discovered sessions","TypeScript or JavaScript knowledge"],"failure_modes":["Requires local access to session log directories — cannot parse logs from cloud-only tools without local caching","Parser accuracy depends on provider log format stability — breaking changes in IDE session formats require parser updates","No real-time streaming of logs — only processes completed session files on disk","Classification is heuristic-based and may misclassify edge cases (e.g., debugging-heavy coding tasks)","Requires sufficient turn context to classify accurately — minimal turns with sparse metadata may default to generic categories","No machine learning model — classification rules are hand-crafted and may not capture domain-specific task patterns","Status reports are point-in-time snapshots — do not track changes over time without manual comparison","Parsing error details may be verbose and difficult to interpret for non-technical users","No automatic remediation — errors are reported but require manual investigation","Plugin development requires understanding the ParsedTurn/Session data model — documentation must be comprehensive","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.5820803657068394,"quality":0.49,"ecosystem":0.7000000000000001,"match_graph":0.25,"freshness":0.75,"weights":{"adoption":0.25,"quality":0.25,"ecosystem":0.1,"match_graph":0.28,"freshness":0.12}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:21.550Z","last_scraped_at":"2026-05-03T13:58:34.540Z","last_commit":"2026-05-03T05:30:48Z"},"community":{"stars":5089,"forks":386,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=getagentseal--codeburn","compare_url":"https://unfragile.ai/compare?artifact=getagentseal--codeburn"}},"signature":"CvI74rWPQHHpMso60mbeLw7BjtCaLNtmmQokBONjwRM8nDWStsZt7OTV2yn4PLG5JEkTR7oHHFKbgPAeWaBkAA==","signedAt":"2026-06-21T08:45:41.142Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/getagentseal--codeburn","artifact":"https://unfragile.ai/getagentseal--codeburn","verify":"https://unfragile.ai/api/v1/verify?slug=getagentseal--codeburn","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"}}