The modern developer terminal has transformed into a high-frequency trading floor for tokens. Within a single session, a programmer might pivot from Claude Code for complex architectural refactoring to Copilot CLI for quick syntax checks, then jump to Gemini CLI to parse a massive codebase. While the productivity gains are immediate, the cost is invisible. Most developers operate in a state of financial blindness, only realizing the scale of their API consumption when the monthly invoice arrives or a quota limit triggers a hard stop. This gap between real-time usage and financial visibility has created a demand for observability tools that treat AI tokens not just as a cost, but as a metric of productivity.
The Mechanics of Token Observability
Show GN enters this ecosystem as a visualization layer designed to bridge the gap between raw logs and actionable insights. The tool does not track usage itself but instead functions as a processor for ccusage, a dedicated AI coding tool usage measurement utility. By reading the local logs generated by ccusage, Show GN can aggregate data from a wide array of AI coding interfaces, including Claude Code, Codex, Gemini CLI, Copilot CLI, and OpenCode. This multi-tool compatibility allows developers to maintain a unified view of their AI spend regardless of which model or provider they are utilizing for a specific task.
The core output of Show GN is the generation of SVG cards that translate abstract token counts into tangible API costs. To ensure these visuals remain relevant to a global audience, the tool integrates a free API to provide real-time exchange rates for four different currencies, including the South Korean Won (KRW). This ensures that the cost reflected on the card is not a static estimate but a current financial reality. To accommodate different aesthetic preferences and layout requirements, the tool offers four distinct card variations: full, half, grass, and combo. Each variation allows the user to choose how much detail to expose, from comprehensive breakdowns to minimalist snapshots.
From Private Logs to Public Proof of Work
What distinguishes Show GN from a standard dashboard is its integration with the social fabric of the developer community. By designing the output as SVG files, the tool enables users to embed their AI usage statistics directly into their GitHub profile READMEs. This transforms a private expense report into a public signal of AI-native development. When paired with automation tools like the Windows Task Scheduler or cron on macOS and Linux, the process becomes a seamless loop. The system reads the logs, generates the updated SVG, and automatically commits the change to the repository daily.
This shift represents a fundamental change in how developers showcase their workflow. For years, the GitHub contribution graph—the famous green squares—has served as the primary proxy for developer activity. However, in an era where a single AI-generated commit can represent hours of manual labor or a few seconds of prompt engineering, the contribution graph is losing its granularity. Show GN introduces a new dimension of proof: the token spend. By visualizing the volume of AI interaction, developers can demonstrate the scale of their AI-assisted orchestration, effectively treating token consumption as a new form of digital craftsmanship.
Because the tool operates entirely within the local environment and does not require external service registrations or the exposure of sensitive API keys, it bypasses the security concerns that typically plague third-party tracking utilities. The tension between the need for visibility and the requirement for privacy is resolved by keeping the data processing local and the output purely visual.
This movement toward AI usage visualization suggests a future where token efficiency becomes a competitive skill, and the ability to orchestrate multiple LLMs is quantified as a core professional competency.
