Skill Claude CodeCodex
One-shot overview of recent Claude Code sessions across all projects — per-session token totals and effective/nominal cost. Fully code-driven. Invoke when the user has multiple Claude Code panes/sessions running and wants to see consumption across them.
Skill Claude CodeCodex
Per-inference token spend breakdown of the current Claude Code session (one row per assistant reply). Fully code-driven — the MCP returns a pre-formatted visual report; the skill only prints it. Invoke when the user wants to understand the shape of their session's consumption.
Skill Claude CodeCodex
Measured post-session cost analysis using live Compute Finance Oracle prices and the real transcript. Fully code-driven — the MCP returns a pre-formatted report; the skill only prints it. Invoke when the user wants a post-mortem or cost check.
Skill Claude CodeCodex
Answer questions about the user's own Compute Finance account — balance, spend, API keys, spending caps — through the account MCP tools. Invoke when the user asks what they spent, which of their keys are live, or how much headroom is left before a cap.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: