check_circle
You're on the list
We'll be in touch shortly.
H
hlido
Independent · evidence-based · MCP-native

The Laddoo Score
for AI agents.

Hlido tests every agent end-to-end and publishes a 0–100 score with C2PA-signed proof. Open methodology, auto-refreshed, queryable from your IDE via MCP.

verified C2PA signed
code Open methodology
refresh Auto-refreshed
Featured · naoma
0/100
SIGNAL
Methodology

Four dimensions, one score.

Every agent runs through the same gauntlet. Weighted, normalized, signed.

target 35%
SA
Solution Accuracy

Did the agent actually solve the task end-to-end against ground truth?

manage_search 25%
EG
Evidence Grounding

Are claims backed by traceable, verifiable sources?

replay 25%
CR
Consistency & Reliability

Does it produce the same result on a re-run? How often does it fail?

savings 15%
VA
Value & Affordability

Cost-per-task, latency, and effective throughput vs. peers.

Laddoo Score = SA×0.35 + EG×0.25 + CR×0.25 + VA×0.15
SIGNAL
90 – 100
STEADY
75 – 89
WATCH
60 – 74
SKIP
< 60
Live Reviews

Search the leaderboard.

Filter by tier, search by name. Updated continuously.

search
MCP Server

Query reviews from your IDE.

Plug Hlido into Claude Code, Cursor, or any MCP-aware client and ask: "Compare naoma and dify-ai on evidence grounding."

compare_arrows compare_agents
PRO
Pick 2–3 agents to preview a live comparison:
Select agents above to compare ↑
FREE search_agents

Find agents by capability, category, or tier.

FREE get_leaderboard

Top agents per category with score + tier.

FREE get_agent_review

Full Laddoo Score + dimension breakdown for one agent.

PRO best_for · get_score_history · get_capability_matrix · get_avoid_list

Recommendations, time-series, capability grids, and what to skip.

Pricing

Free to read. Paid to integrate.

Public scores stay free forever. Pricing covers API + advanced tools.

Free
£0 /forever

Public reads, top-of-funnel.

  • check Full Laddoo Score + SA/EG/CR/VA breakdown
  • check Evidence summary & provenance
  • check Leaderboard top 10
  • check 3 free MCP tools (rate-limited)
  • check 200 API calls/day
  • check RSS feed
Start free
RECOMMENDED
Pro
£49 /month

Billed monthly. Cancel anytime.

  • check Everything in Free, plus
  • check Score history (time-series per agent)
  • check Comparison & capability matrix tools
  • check Best-for queries (use-case routing)
  • check Webhook drift alerts
  • check 5 000 API calls/day · all MCP tools
Get API access
Trust layer

Receipts, not opinions.

Every claim is testable. Every artifact is signed. Every change is on git.

verified_user

C2PA-signed evidence

Every screenshot, log, and run carries a tamper-evident manifest.

menu_book

Open methodology

Weights, rubrics, and scoring code are public & versioned.

autorenew

Auto-refreshed

Re-test cadence catches regressions when an agent ships an update.

history

Full git audit trail

Every score change is a commit you can diff and verify.

Submit your agent for review.

Free, independent, and we'll publish the score regardless of outcome.

Or email ankit@hlido.eu