Skip to main content

Discover, analyze, and optimize your prompts from AI coding sessions

Project description

re:prompt

Grammarly for Prompts -- analyze your AI conversations with research-backed scoring.

PyPI version Python 3.10+ License: MIT Tests Coverage


See it in action

$ pip install reprompt-cli
$ reprompt scan --source claude-code
  Scanning 890 sessions... Found 1,063 prompts (295 unique)

$ reprompt score "Fix the auth bug in src/login.ts where JWT expires"
  Score: 74/100
  Structure: 18/25 | Context: 22/25 | Position: 15/20 | Repetition: 9/15 | Clarity: 10/15
  Tip: Add the error message for +15% accuracy

$ reprompt wrapped --share
  Your Prompt DNA: Architect (Score 78, better than 72% of prompters)
  Share link: https://getreprompt.dev/w/abc123

What it does

Command Description
reprompt scan Scan 8+ AI tools for prompts (Claude Code, Cursor, Aider, Gemini CLI, Cline, OpenClaw, ChatGPT, Claude.ai)
reprompt score Instant 0-100 scoring with 30+ research-backed features
reprompt compare Side-by-side analysis of two prompts
reprompt wrapped Your Prompt DNA report -- persona, scores, shareable card
reprompt insights Personal patterns vs research-optimal
reprompt digest Weekly summary comparing current vs previous period
reprompt report Full analytics with hot terms, clusters, patterns
reprompt library Auto-extracted prompt patterns and templates

Prompt Science

Scoring is based on 4 research papers (Google, Stanford, EMNLP, Prompt Report) covering 30+ features across 5 dimensions:

Dimension What it measures
Structure Markdown formatting, code blocks, explicit constraints
Context File paths, error messages, technical specificity
Position Instruction placement relative to context
Repetition Redundancy that degrades model attention
Clarity Readability, sentence length, ambiguity

All analysis runs locally in <1ms per prompt. No LLM calls, no network requests.

Supported AI tools

Tool Format Session location
Claude Code JSONL ~/.claude/projects/
Cursor .vscdb ~/Library/Application Support/Cursor/User/
Aider Markdown .aider.chat.history.md
Gemini CLI JSON ~/.gemini/tmp/
Cline (VS Code) JSON globalStorage/saoudrizwan.claude-dev/
OpenClaw / OpenCode JSON ~/.openclaw/ / ~/.opencode/sessions/
ChatGPT JSON conversations.json export
Claude.ai JSON/ZIP Web chat export

Installation

pip install reprompt-cli          # core
pip install reprompt-cli[chinese] # + Chinese prompt support
pip install reprompt-cli[mcp]     # + MCP server for Claude Code / Continue.dev / Zed

Privacy

  • All analysis runs locally. No prompts leave your machine.
  • Optional telemetry sends only anonymous 26-dimension feature vectors -- never prompt text.
  • Open source: audit exactly what's collected.

Privacy policy

Links

Contributing

See CONTRIBUTING.md for development setup and guidelines.

License

MIT

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

reprompt_cli-1.2.0.tar.gz (1.0 MB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

reprompt_cli-1.2.0-py3-none-any.whl (217.3 kB view details)

Uploaded Python 3

File details

Details for the file reprompt_cli-1.2.0.tar.gz.

File metadata

  • Download URL: reprompt_cli-1.2.0.tar.gz
  • Upload date:
  • Size: 1.0 MB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.7

File hashes

Hashes for reprompt_cli-1.2.0.tar.gz
Algorithm Hash digest
SHA256 128581c2eaafd1530a7cc3c9b3c1209771371f6aa841db26e3c444cb6aa4926a
MD5 5b6b9017e601b6ec770eddf9d14d52bb
BLAKE2b-256 25127b87a999eebf958210ffb0cc89a706150de4a55bd907df5098e2da067fd2

See more details on using hashes here.

File details

Details for the file reprompt_cli-1.2.0-py3-none-any.whl.

File metadata

  • Download URL: reprompt_cli-1.2.0-py3-none-any.whl
  • Upload date:
  • Size: 217.3 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.7

File hashes

Hashes for reprompt_cli-1.2.0-py3-none-any.whl
Algorithm Hash digest
SHA256 409a71497853035d94f69f98f6e8b9b6b0d427bc5dba67848024831de425f061
MD5 7563401bc3b94833a751ed3d7c6a936d
BLAKE2b-256 f8ec27d22bb79b426bad26c3e3fc0bc00b6d6a76cbea81ecc5930381ced9a910

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page