agent-cost
Report what Claude Code work costs at API list prices, broken down by model, project, skill, tool and token type, from session transcripts or OpenTelemetry logs, and name the changes most likely to reduce it. Use when someone asks what their agents cost, why usage is so high, which skill, tool or project is most expensive, or what switching model would save. Not for reducing spend in code that calls the Claude API - that is claude-api.
- Compatibility
- Requires python3 (3.9 or later, standard library only) and Claude Code session transcripts under ~/.claude/projects. OpenTelemetry logs are optional.
Pinned to revision 6afcca115bdd, so it is the text this page describes rather than whatever the author pushed since.
Pre-approved tools experimental
Experimental field. Support varies between clients, so this list is what the author declared, not what your client will enforce.
- Bash(python3:*)
- Bash(ls:*)
- Read
Files
Every link opens the file at its source, pinned to the revision this page describes.