textfsm-ai

AI-powered TextFSM template generation, parsing assistance, and smart log extraction.

textfsm-ai brings modern LLM intelligence to traditional TextFSM workflows.
It helps you automatically generate templates, validate patterns, explain parsing logic, and accelerate network automation development.

🚀 Features

• AI-Powered Template Generation — Turn raw CLI output into production-ready TextFSM templates in seconds.
• Smart Validation & Refinement — Automatically verify template correctness and refine ambiguous patterns with AI assistance.
• 18 Supported LLM Providers — Pick the provider and model for each task; every provider is called through one consistent interface.

📦 Installation

pip install textfsm-ai

This installs the core CLI/API only. Add the provider(s) you actually use as an extra, e.g. pip install textfsm-ai[openai] or pip install textfsm-ai[anthropic] — or textfsm-ai[all] for every provider SDK at once. See Installation for the full list.

Verify installation

textfsm-ai --version
# or
textfsm-ai version

Either command prints the installed version, e.g. textfsm-ai v0.4.0.

📚 Documentation

• Latest docs: https://geeks-trident-llc.github.io/textfsm-ai/latest/
• All versions: https://geeks-trident-llc.github.io/textfsm-ai/

🤖 What is textfsm-ai, in an LLM's view?

For an LLM, TextFSM's native syntax — regex-heavy Value/Rule definitions driving a state machine — is easy to get almost right and hard to get exactly right; small regex or state-transition mistakes are common and easy to miss. textfsm-ai acts as an LLM optimizer for TextFSM: instead of asking a model to freehand raw TextFSM syntax, it gives the model a smaller, constrained, readable DSL to generate from, then deterministically compiles that DSL into a canonical TextFSM template. The model reasons about structure and intent; textfsm-ai guarantees the correctness and consistency of the resulting syntax.

❓ Why do you need textfsm-ai?

• Understands Messy Input — Feed it raw CLI output, log lines, or any plain-text or semi-structured text; no need to hand-write parsing rules for every format and edge case.
• No TextFSM Syntax Required — textfsm-ai handles TextFSM's regex-heavy Value/Rule/state-machine syntax for you, so you don't need prior TextFSM expertise to get a working, correct template.
• Canonical, Consistent Templates — Every generated template goes through the same deterministic normalization step, so the same kind of input reliably produces the same shape of output every time.
• Readable DSL — Templates can also be expressed as a human-readable DSL that non-technical teammates can read and reason about, instead of a wall of regex.
• Recognizer Pattern Generation — Beyond extracting values, textfsm-ai can generate recognizer patterns that detect which block of text a template applies to before parsing it.
• Real-World Applications — Log extraction, test-data extraction and normalization, and other pipelines that turn recurring plain-text or semi-structured output into reliable, structured records.