
Description
receipts is an agent skill from Anthropic's official plugins collection, listed on skills.sh. It generates a personal Claude Code usage and impact report, called receipts, from the local session transcripts on the machine. It is for justifying Claude Code usage or spend to someone else. The data is every session in the transcript folder under ~/.claude/projects, already on disk and needing no set-up. The mining step is a local Node script that uses file reading and regular expressions, with no API calls. The only model call is one final write-up over a small JSON summary, however much history was scanned. To check the picture, it cross-references the local git log for each repository, so commit activity can be compared with Claude Code session activity, all without network access. The first step is to work out the reporting period. The output is a markdown report of one developer's own activity. It suits individual developers, team leads or finance reviewers who want evidence of what Claude Code has been used for.…
Highlights
- Built from local session transcripts
- No API calls while mining the data
- Cross-checks git commits per repository
- Markdown report of your own activity
Ratings & Reviews
No ratings yet
Be the first to rate receipts.
People also visit

timeline-report
Generate a "Journey Into [Project]" narrative report analyzing a project's

content-research-writer
Assists in writing high-quality content by conducting research

batch-grill-me
A relentless interview that asks every frontier question at once, round by round

coding-agent
Delegate coding work to Codex, Claude Code, or OpenCode as background workers
receipts FAQ
What does the receipts skill produce?
A markdown report of your own Claude Code usage and impact, built from the local session transcripts on your machine.
How do I install it?
Run: npx skills add https://github.com/anthropics/claude-plugins-official --skill receipts
Does it send my transcripts anywhere?
The mining step runs locally with no API calls. The only model call is one write-up over a small summary, according to the skill.
Where does the data come from?
The session transcripts in ~/.claude/projects, plus a local git log for each repository.
Has it been checked for security?
skills.sh lists three audits: Gen Agent Trust Hub Pass, Socket Pass and Snyk Pass.
What are the best receipts alternatives?
Popular alternatives to receipts include session-report, timeline-report and analyze. See all receipts alternatives