Tesseract OCR
Skill8 starsChanged 7 months agoApache-2.0View on GitHub
Extract text from images and scanned PDFs via the local Tesseract binary.
- Installs packages
In this repository38 other files it gives its agents.
- Brave SearchSkill
- BrowserSkill
- CloudflareSkill
- DNS ToolsSkill
- Docker APISkill
- Endpoint CheckSkill
- Fetch URLSkill
- GitHubSkill
- Google OAuthSkill
- GoogleSkill
- GoPlacesSkill
- LoopSkill
- Microsoft Edge TTSSkill
- 1PasswordSkill
- OpenAI WhisperSkill
- OpenStreetMapSkill
- PDF ExtractSkill
- ProxmoxSkill
- RememberSkill
- RSS / AtomSkill
- RTSP SnapshotSkill
- Session LogsSkill
- SimplifySkill
- SkillifySkill
- SpotifySkill
- SSH ExecSkill
- StuckSkill
- SummarizeSkill
- TaskFlowSkill
- tmuxSkill
- TranslateSkill
- Update ConfigSkill
- VerifySkill
- Video FramesSkill
- WeatherSkill
- WikipediaSkill
- Wolfram AlphaSkill
- yt-dlpSkill
The file
Files
Tesseract OCR
Local, offline OCR. Wraps the tesseract CLI. No API key, no upload.
Use when
- "Read the text in this screenshot / photo"
- Extracting text from scanned documents (PDF images)
- Processing receipts / handwritten notes (low confidence, but tries)
- Pre-processing an image before feeding text into
summarize/translate
Do not use when
- Digital PDFs with embedded text — use
pdf-extract(faster, exact) - Handwriting-heavy or complex layouts — results will be poor
- You need bounding boxes or structured output — use a vision LLM
Tools
status— Tesseract versionlanguages— installed language packs (eng,spa,fra, ...)ocr—image_path; optionallang(defaulteng),psm(page seg mode, default 3),oem(engine mode, default 3)
Language packs
Install on Debian/Ubuntu: apt install tesseract-ocr-spa (etc.). Check
installed with languages. Combine with +: lang: "eng+spa".
Common PSM values
| PSM | Meaning |
| 3 | Fully automatic (default) |
| 6 | Single uniform block of text |
| 7 | Single line |
| 11 | Sparse text — find as much as possible, no order |
| 13 | Raw line — single text line, no preprocessing |
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.