V26.09:crwdla/tokentab

Overview

Tokentab addresses a genuine pain point: developers using multiple AI coding assistants (Claude Code, OpenAI Codex, Gemini CLI) have no unified view of their spending. Each platform reports costs separately, if at all, making it hard to understand total AI expenditure.

Value proposition:

Multi-platform cost aggregation is genuinely useful — no single tool currently does this across Claude, Codex, and Gemini
CLI-first approach fits the developer workflow perfectly
"By project and day" breakdown is actionable for budgeting and cost optimization
Low friction adoption: reads existing log files, no API integrations needed
Clear monetization paths: open-source core with pro features (CSV export, alerts, dashboards) or a hosted dashboard

The timing is excellent — AI coding tool usage has exploded, and cost visibility is a top concern for individuals and teams.

Score Breakdown

What They Do

A CLI that reads Claude Code, Codex, and Gemini CLI session logs and works out how much they cost, by model, project, and day.

Why It Matters

Value proposition:

Multi-platform cost aggregation is genuinely useful — no single tool currently does this across Claude, Codex, and Gemini
CLI-first approach fits the developer workflow perfectly
"By project and day" breakdown is actionable for budgeting and cost optimization
Low friction adoption: reads existing log files, no API integrations needed
Clear monetization paths: open-source core with pro features (CSV export, alerts, dashboards) or a hosted dashboard

The timing is excellent — AI coding tool usage has exploded, and cost visibility is a top concern for individuals and teams.

Business Model

To be determined based on project analysis.

CRP Decision Report

🧠 Advisor

Value proposition:

Multi-platform cost aggregation is genuinely useful — no single tool currently does this across Claude, Codex, and Gemini
CLI-first approach fits the developer workflow perfectly
"By project and day" breakdown is actionable for budgeting and cost optimization
Low friction adoption: reads existing log files, no API integrations needed
Clear monetization paths: open-source core with pro features (CSV export, alerts, dashboards) or a hosted dashboard

The timing is excellent — AI coding tool usage has exploded, and cost visibility is a top concern for individuals and teams.

[score: 8]

[score: 8.0]

🧨 Devil

Risk 1: Log format fragility. Each platform could change their log format without notice. Claude Code's log structure, Codex's output format, and Gemini's session files are all internal implementation details. A single update from any vendor could break the parser, requiring constant maintenance.

Risk 2: Minimal defensibility. This is essentially a log parser with formatting. There are no network effects, no data moat, no switching costs. Anyone can read the same log files. A competitor (or even the platforms themselves) could ship this feature natively at any point, obliterating the product.

Risk 3: Narrow market & low willingness to pay. Individual developers using CLI tools are famously frugal. The people most motivated to track costs are also the least likely to pay for a tool. Enterprise teams (who would pay) may not use these CLIs in ways that generate accessible logs.

Risk 4: Pricing complexity. Token counts ≠ dollar costs. Rate cards vary by model version, region, and contract. The tool either oversimplifies pricing (losing accuracy) or requires constant rate card updates (maintenance burden).

[score: -6]

[score: -6.0]

📚 Historian

Similar projects in the "developer tooling for AI" space show mixed results:

Aider's cost tracking (built-in, ~2024): Open-source tools that bake in cost visibility tend to get abandoned as a feature once the core team moves on. It was deprioritized quickly.
GitHub's Copilot billing dashboards: Platform-native cost tracking tends to kill third-party alternatives. When Anthropic, OpenAI, and Google each build cost dashboards into their own platforms, the incentive for a cross-platform tool shrinks.
Terminal-based CLI tools (e.g., glam, zoxide, dua): The pattern is strong community adoption but weak monetization. The "it's free on GitHub" mentality is intense in this segment.
Token/usage trackers (e.g., for ChatGPT API): Niche tools that solved a real problem but never scaled beyond thousands of users. They survived as side projects, not businesses.

The favorable pattern: CLI tools for developers can gain traction fast (think gh, starship, lazygit). The dangerous pattern: no moat against platform-native solutions and extremely low willingness to pay.

[score: 1]

[score: 1.0]

🧮 Budget Steward

Development cost estimate (solo indie dev):

Rust/Go CLI implementation (log parsing for 3 platforms): ~3,000–5,000 equivalent in dev time or outsourced
Testing across platform versions: ~$500
CI/CD setup (GitHub Actions): ~$0 (free tier)

Infrastructure cost (3 months):

GitHub repository: $0
No backend needed (pure CLI, reads local files): $0
Documentation site (GitHub Pages): $0

Third-party services:

No API calls required (reads local logs): $0
Homebrew/aur/package maintainers: $0

Compliance:

Open-source license: $0
No data collection = no privacy compliance needed

Total estimated cost: ~3,500–5,500 (mostly dev time, near-zero out-of-pocket)

This is well under the 15,000 threshold. Even outsourcing key components would likely stay under 10,000. The funding gap is negative — this is highly budget-friendly.

[score: +1]

🧱 Founder

Should we build this? Yes, but with scoped expectations.

This is a strong side-project or community tool play, not a venture-scale business. The budget is tiny, the pain point is real, and the market is growing. But the ceiling is low.

Recommended approach:

Launch as open-source (MIT) CLI to build audience and credibility fast
Add a "Pro" tier later: hosted dashboard, team sharing, cost alerts, CSV exports
Target: 1,000+ GitHub stars in 3 months, 100+ weekly active users
Timeline: MVP in 4–6 weeks, first paying customers in 3–6 months via the pro tier

The ask: Can you accept that this might be a 5K–20K/year business rather than a startup? If yes, it's a smart, low-risk play. If you need venture-scale returns, look elsewhere.

[score: 7]

[score: 7.0]

Why This Made Only 10

This project was selected because it scored highly across our evaluation framework. The CRP analysis confirmed strong potential, making it one of the most promising opportunities this week.