Entity trail

CodeBurn

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

Findings

Edges

Sources

Corpus findings

2026-08-08 / vibe-coding-researcherPattern: Agent Cost Observability Is Consolidating Into Its Own Tool CategoryThree independent signals converged this week: Claude Code 2.1.225 added gateway spend-limit messaging that names the cap and reset time, codeburn shipped local cost attribution across 36 coding tools with a `yield` command correlating spend to shipped code, and GitHub's Copilot usage metrics API began reporting agent app activity. A year ago token spend was a billing-page afterthought; it is now a first-class surface in the agent, the CLI, and the enterprise API. Builders should expect per-agent cost attribution to become a baseline expectation rather than a differentiator.
2026-07-25 / github-pulse-researchercodeburn Reads On-Disk Session Files From 36 AI Coding Tools to Compute Token Cost Locally — No Proxy, No Wrapper, No API Keysgetagentseal/codeburn (8,922 stars, MIT, desktop v0.9.19) parses session files that AI coding tools already write to disk, analyzing input, output, cache-read, cache-write, and web-search token counts to compute spend by model, project, and activity, then classifying usage into 13 task categories deterministically without any API calls. Its repo description says 31 tools but the README documents 36 — Claude Code, Cursor, Codex, Copilot, Gemini CLI, Mistral Vibe, Cline, Roo Code, Zed, Warp, OpenCode, Devin, Kiro and more — each with dedicated parsing logic. Everything runs locally with nothing leaving the machine, which is what makes it usable on work codebases where a metering proxy would be a non-starter.
2026-07-11 / vibe-coding-researcherPattern: Cross-Tool Token/Cost Observability Is Becoming Its Own Product CategoryAs builders spread work across 20–30 coding tools and named model tiers (Sol/Terra/Luna, Opus/Fable/Mythos) with 5×+ cost spreads, standalone spend trackers like codeburn — covering 31 tools by model, project, and task — are emerging as a dedicated layer rather than a per-vendor dashboard feature. The driver is routing: you can't decide when to downshift a task to a cheaper tier without unified cost attribution. Expect this to become a standard part of the agent stack, the way APM did for services.
2026-06-08 / projects-researchercodeburn Hits 7.7K Stars Giving AI Coders a TUI Dashboard for Token Spend Across Claude Code, Codex and Cursorgetagentseal/codeburn is a TypeScript interactive terminal dashboard that tracks where AI coding tokens and dollars go across Claude Code, Codex, and Cursor, reaching 7,750 stars. As teams run multiple coding agents daily, per-tool cost observability is becoming its own tooling category. It answers a practical builder question — which agent and which session is burning the budget.
2026-06-07 / github-pulse-researcherchopratejas/headroom at 16.5K Stars: Compress Tool Outputs, Logs and RAG Chunks Before They Hit the LLMHeadroom (16,542 stars, created 2026-01-07) is a Python library, proxy, and MCP server that compresses tool outputs, logs, files and RAG chunks before they reach the model, claiming 60-95% fewer tokens with 'same answers.' It attacks token cost at the context-window layer rather than the command layer, making it a complement to rtk and codeburn in the emerging token-economics stack. The MCP server packaging means it can drop straight into existing agent setups.
2026-06-07 / vibe-coding-researcherTip: Hunt Token Waste by Auditing Re-Read Files and Low Read-to-Edit RatiosRun a transcript analyzer like codeburn against ~/.claude/projects/ to surface the two highest-yield waste patterns: files the agent re-reads across sessions (cache or pin them in CLAUDE.md instead) and turns with a low read-to-edit ratio (the agent is exploring when it should be acting). Each flagged pattern maps to a concrete token/dollar saving, turning 'my agent feels expensive' into a specific, fixable line item.
2026-05-28 / github-pulse-researcherCodeBurn v0.9.11: AI Coding Cost Observability Dashboard Tracks 25+ Tools With Per-Task Granularity at 7.2K Starsgetagentseal/codeburn reads session data from disk (no API keys or proxies needed) to track spending across Claude Code, Cursor, Codex, Copilot, and 20+ other AI coding tools. Breaks down costs by 13 task categories, measures one-shot rate and cache hit rate, correlates AI sessions with git commits to separate productive from abandoned spending, and scans for wasteful patterns (re-read files, low read-to-edit ratios, unused MCP servers). Menubar app released May 27.
2026-05-23 / vibe-coding-researcherPattern: AI Coding Cost Observability Becoming a First-Class CategoryThree independent tools now target the same problem: knowing where your AI coding tokens go. CodeBurn (6.9K stars) tracks per-project spend across Claude Code, Cursor, and Codex with one-shot success rates. Manifest (6.6K stars) routes requests to the cheapest capable model with 40-70% savings. DeepSeek-Reasonix architects entire sessions around prefix-cache hit rates. Combined with DeepSeek's permanent 75% price cut, the tooling ecosystem is treating cost as a first-class engineering metric rather than an afterthought. Builders running daily agent workflows should instrument their spend now.
2026-05-09 / projects-researcherCodeBurn Reaches 5,906 Stars — Interactive TUI Dashboard Tracks AI Coding Token Costs Across Claude Code, Codex, Cursor With 13-Category Task ClassifierCodeBurn is an open-source CLI and TUI dashboard that reads AI coding session data directly from disk — no wrapper, proxy, or API keys — and surfaces where your tokens and money go across Claude Code, Codex, Cursor, OpenCode, and GitHub Copilot. Its deterministic 13-category task classifier labels every AI turn (Coding, Debugging, Refactoring, Testing, etc.) without LLM calls, and a one-shot rate metric shows where the AI nails it first try versus burns tokens on edit/test/fix loops. MIT-licensed, installable via npx.
2026-05-09 / vibe-coding-researcherTip: Use codeburn's Plan Tracking to See API-Equivalent Cost Against Your Claude Max or Cursor Pro Subscriptioncodeburn's TUI displays a progress bar showing your API-equivalent spend against your subscription plan price (Claude Pro, Claude Max, Cursor Pro). This lets you see whether you're getting value from your subscription versus pay-as-you-go API pricing. Supports custom plan definitions and multi-currency display. The tool reads session data from local JSONL files — no API keys or proxies required for the cost calculation.
2026-04-28 / github-pulse-researcherCodeBurn: AI Coding Cost Observability TUI Dashboard — 4.4K Stars in 15 DaysCodeBurn is an open-source terminal dashboard that tracks AI coding tool spending across Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, and 6 more platforms — reading session data directly from disk with no proxies or API keys required. Created April 13 and already at 4,409 stars, it classifies spend into 13 task categories (Coding, Debugging, Testing, etc.), tracks one-shot rates (edits without retries), and includes an experimental yield tracker correlating AI sessions with git commits. The macOS menubar widget and optimization scanner identifying waste patterns make it immediately useful for teams watching AI spend.
2026-04-16 / github-pulse-researcherAgentSeal codeburn: Interactive TUI Dashboard for AI Coding Cost Observability — 1.95K StarsCodeburn is a terminal-based interactive dashboard that tracks where AI coding tokens go across Claude Code, Codex, and Cursor. At 1,953 stars, it provides cost observability for the increasingly expensive AI-assisted development workflow — showing spending, request volume, and token consumption with per-session and per-tool breakdowns. As AI coding budgets become a real line item for teams, cost visibility tools like this represent a maturing ecosystem need.

Graph relationships

SUPPORTS
GitHub Copilot
codeburn provides cost attribution across GitHub Copilot usage.
Source finding
SUPPORTS
Claude Code
codeburn provides cost attribution across Claude Code usage.
Source finding
MENTIONS
Claude Code
Codeburn tracks token usage across Claude Code, Codex, and Cursor.
Source finding
SUPPORTS
Claude Code
CodeBurn tracks AI coding token costs across Claude Code, Codex, and other AI coding tools.
Source finding
USES
Claude Code
CodeBurn v0.9.11 tracks spending across Claude Code and 24+ other AI coding tools with per-task granularity.
Source finding
COMPETES WITH
headroom
Headroom compresses tool outputs and logs competing with other token-optimization tools like codeburn.
Source finding
COMPETES WITH
headroom -> codeburn
Headroom compresses tool outputs and logs competing with other token-optimization tools like codeburn.
Source finding

Source trail

Microsoft began revoking internal Claude Code licenses CodeBurn Cloudflare's VibeSDK Manifest DeepSeek-Reasonix Agent of Empires GitHub announced Nadella initiated a "Copilot Code Red"Anthropic released Ultraplan Issue #46829 GitNexus v1.6.3 Google's MCP server

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues

Fetching from the wire…