giskard-oss
5,295Open-source evaluation and testing library for LLM agents and ML models.
DetectionPost-generationFull HITLStableTier 2AEnvironment
Maturity tier details
Computed tier: 2A(raw score 4.99). Effective tier: 2A.
Facets:documented_in=0evaluated_on=0integration_surfaces=1.5paper_citations=0recent_commit=2stars=1.49
Source of truth:pipeline/maturity/score.py
Description
Giskard is an open-source testing library originally for tabular ML, extended to cover LLM apps and agents. Runs pre-defined robustness, bias, and hallucination tests and generates regression reports. Used in CI to gate ML/LLM deployments.
Why it's included
ML-lineage evaluation cell: extends classical ML testing primitives to LLMs, differentiating from LLM-native tools like deepeval.
Addresses failure modes
- Security vulnerability — Insecure patterns, injection flaws, weak cryptography, secret leakage.
- Fabrication — Generating reference to non-existent APIs, packages, or symbols.
- Logic error — Correct API use, incorrect logic.
- Scope creep — Actions outside declared task boundary.
- Test manipulation — Agent gaming its own evaluation — editing tests or test harness so failures disappear without fixing the underlying bug.
Techniques
robustness-testsbias-testshallucination-testsci-gates