An Open-Source End-to-End Pipeline for Multi-Planet Detection and Bayesian Characterisation in Kepler Long-Cadence Photometry: Validation on Kepler-11

We present an open-source, end-to-end pipeline for the detection and Bayesian characterisationof multiple transiting exoplanets in Kepler long-cadence photometry. The pipeline integrates (i) aSavitzky–Golay detrending and sigma-clipping preprocessor, (ii) Box Least Squares (bls) for rapidperiod scanning, (iii) an iterative Transit Least Squares (tls) multi-planet search with physically motivated transit templates and a calibrated Signal Detection Efficiency (SDE) statistic, (iv) automatedcross-matching against the Kepler Objects of Interest (KOI), confirmed-planet, and false-positive catalogs, and (v) a joint Metropolis–Hastings Markov Chain Monte Carlo (mcmc) fitter that simultaneously constrains the orbital period P, mid-transit time t0, transit depth δ, duration τ , and quadraticlimb-darkening coefficients (u1, u2) for all detected planets. We validate the pipeline on a syntheticphotometric realisation of Kepler-11, a six-planet compact system with transit depths spanning 310–1630 ppm and orbital periods 10.3–118 days. All five planets recoverable within the four-year baselineare detected at SDE ≥ 7.8; tls recovers 5/5 planets compared with 3/5 for bls at the same falsealarm threshold. Joint mcmc fitting yields period uncertainties σP /P < 0.03% and depth uncertaintiesσδ/δ < 10% for all detected planets. The pipeline is implemented in pure Python with minimal dependencies and is designed for portability across both locally cached and MAST-hosted data.