---

name: flashtext description: "Routes FlashText tasks for keyword extraction, replacement, fuzzy matching, keyword loading, and keyword-inspection workflows." disable-model-invocation: true metadata: disco-role: operating license: MIT

FlashText

Use this skill when a request names FlashText or asks to extract, replace, load, inspect, or fuzzily match keywords with KeywordProcessor.

This repository is small enough that one router covers the whole package. Read the bundled references for exact API details instead of reopening the original checkout.

Install and verify

• Public install: python -m pip install flashtext
• From a checkout: python -m pip install -e .
• Safe smoke check: python scripts/check_install.py
• JSON smoke output: python scripts/check_install.py --json
• Minimal import check: python -I -c "from flashtext import KeywordProcessor; print(KeywordProcessor().__class__.__name__)"

If you are checking whether this skill still matches the repository snapshot, read references/repo-provenance.md before making that call.

Route map

• Need to add, remove, count, or inspect keywords with dictionary-like behavior (len, in, [], get_keyword, get_all_keywords)? Read references/api-reference.md and references/workflows.md.
• Need sentence extraction, replacement, spans, case_sensitive, or fuzzy max_cost behavior? Start with references/workflows.md, then open references/api-reference.md for exact signatures and return shapes.
• Need to load keywords from a file, list, or dictionary, or you want the expected input shapes? Read references/data-formats.md.
• Need to debug import errors, bad file paths, malformed list/dict shapes, replace_keywords with tuple clean names, or missed matches caused by word boundaries or case handling? Read references/troubleshooting.md, then rerun scripts/check_install.py.
• Need to compare the current checkout against the recorded baseline? Read references/repo-provenance.md.

Typical request phrases

• "extract keywords from this sentence"
• "replace keywords with clean names"
• "show spans for each match"
• "make matching case sensitive"
• "load keywords from a file"
• "load from a list or dictionary"
• "remove a keyword variant"
• "allow punctuation like / inside keywords"
• "debug why a keyword did not match"
• "check whether the package is installed correctly"

What to read or run next

• Use references/workflows.md for step-by-step usage patterns.
• Use references/api-reference.md when you need the exact signature or return type for a method on KeywordProcessor.
• Use references/data-formats.md when shaping file, list, or dictionary inputs.
• Use references/troubleshooting.md for predictable input and matching failures.
• Run scripts/check_install.py whenever you need a quick install/import/ smoke verification.

What this skill does not cover

• No CLI entry point is shipped by the package.
• No network service, model backend, or accelerator-specific workflow is involved.
• No training, benchmarking, or docs-build workflow is part of the runtime route.

Working rule

Keep runtime instructions self-contained. Use the bundled references and script instead of the original repository when you need API details, data-shape reminders, or a fast validation check.