AI agent glossary
Plain-English definitions of the terms behind independent AI-agent evaluation — each backed by 794 hands-on reviews.
- Laddoo Score — The Laddoo Score is Hlido’s 0–100 rating for an AI agent, set by hands-on, claim-by-claim …
- VITAL tier — VITAL is Hlido’s top tier — an AI agent scoring 90 or above. It means the agent genuinely …
- STEADY tier — STEADY is Hlido’s middle tier (score 70–89): a genuinely useful AI agent that works, with …
- FADING tier — FADING is Hlido’s warning tier (score 40–69): the demo works, but the real workload expose…
- FLATLINE tier — FLATLINE is Hlido’s lowest tier (score below 40): the AI agent fails to deliver on its cor…
- AI agent claim verification — Claim verification is mapping every marketing claim an AI agent makes to PASS, FAIL, or UN…
- independent AI agent review — An independent AI agent review is an evaluation run by a third party that takes no money f…
- evidence-backed review — An evidence-backed review ties every verdict to a specific hands-on run on a specific date…
- AI agent drift — Agent drift is the decay of an AI agent’s quality over time as models change, prompts rot,…
- AI agent reliability — AI agent reliability is whether an agent delivers consistently under real, messy workloads…
- AI agent trust score — An AI agent trust score is a single, independent signal of how much an agent can be relied…
- machine-readable agent review — A machine-readable review publishes its verdict as structured data — a signed scorecard, a…