Robot Framework Failure Summary

A Robot Framework listener that mutes noisy engine logs (Selenium/Browser internals) and generates a clean, interactive failure summary report — with deep links to the failing keyword in log.html and IDE-ready file:line locations you can paste into VS Code (Ctrl+P) or PyCharm (Ctrl+Shift+N).

Works with SeleniumLibrary and Browser library, and with both robot and pabot.

Install

pip install robotframework-failuresummary

For parallel runs: pip install robotframework-pabot

Usage

Plain robot (single process)

these tests can be anything like relative path to ur .robot file or u can change the cd and run it

rf-failure-summary tests 

or, if you don't want the wrapper and just want plain robot directly (this also works fine, no pabot involved):

robot --listener RobotFailureSummary.listener --outputdir results tests

Pabot (parallel execution)

rf-failure-summary tests --pabot --processes 4

⚠️ For pabot, you must use rf-failure-summary (not plain pabot --listener ...) — see "Why not just --listener?" below for why.

Other options (work with both)

rf-failure-summary tests --outputdir results             # custom output folder
rf-failure-summary tests --include smoke                  # pass tags/flags through to robot/pabot
rf-failure-summary tests --pabot --processes 4 --include smoke --exclude wip

Open results/failure_summary.html after the run finishes.

Why not just --listener? Under pabot, results from all worker processes only get merged after they've exited — a listener can't hook into that. rf-failure-summary runs robot/pabot itself, waits for the full run (merge included), then generates one complete report. Plain robot --listener RobotFailureSummary.listener still works fine on its own since a single robot process can generate the complete report by itself.

What you get

Each failure card shows the classification (Assertion/Timeout/Browser/Other), the keyword path with file:line for each step, the failure message, and a Jump to Failing Keyword button into log.html.

Config

Debug logging

If a failure card shows file.robot:? instead of a real line number, enabling debug logging won't fix it — it just records what the tool tried while looking up that keyword, so you can attach listener_debug.log when reporting it as a bug (or dig into it yourself if you're comfortable).

Windows (PowerShell):

$env:ROBOFAILURESUMMARY_DEBUG="1"
rf-failure-summary tests

Windows (cmd):

set ROBOFAILURESUMMARY_DEBUG=1
rf-failure-summary tests

macOS / Linux:

export ROBOFAILURESUMMARY_DEBUG=1
rf-failure-summary tests

This is off by default — leave it unset for normal use, since it writes a log file on every run.

Changelog

2.0.0.1 - Updated readme

2.0.0 — Added rf-failure-summary base 1.0.0 features + CLI with pabot support, included line-number resolution across multi-file suites, responsive UI with failure seperation based on label.

1.0.0 — Initial release. deep linking failure nodes navigation in a single click which keeps that particular failure node open and rest all closed for easier and optimized log.html navigation for best usage u need to keep both the base html report tab and failure_summary.html tab in single window as clicking to failure nodes first time opens new tab second time it reuses the already opened tab thats why single window spliting tabs will give you the best output.

License

MIT