UI-TARS
Pioneering Automated GUI Interaction with Native Agents
Is UI-TARS safe? Thin or incomplete trust evidence. Review carefully before production use.
Compare UI-TARS
How does it stack up against its Browser & Computer Use neighbours?
Pick any agent to compare →In detail: UI-TARS scores 41.0/100 (Grade D), ranked #1172 of 1371 tracked open-source AI agent projects, on evidence coverage A (4 of 5 independent signal types). The public evidence: no package-provenance attestation found; OSSF Scorecard rates its supply-chain practices 2.5/10; 19% of recent commits are signed; no published advisory affects its latest release (ui-tars); last pushed 2026-01-27. Every point is earned from checkable signals — never paid placement. How scoring works →
How UI-TARS could raise its score
Each line changes one public signal and recomputes with the live scoring function. The gains don't add up exactly, because the score is capped near the top.
- Publish build provenance for its packagesPyPI Trusted Publishing attestations +14.7 → 55.7 C · #25 of 45
- Raise the OSSF Scorecard from 2.5 to 9.0lowest checks: Branch-Protection 0, CI-Tests 0, CII-Best-Practices 0 +13.6 → 54.6 C · #25 of 45
- Sign every commit19% signed today +4.0 → 45.0 D · #35 of 45
Chain of custody
Scanners check what the code says. This traces who ships UI-TARS and whether that has changed: from the source repository, through how changes are reviewed and released, to the package you install. These are the checks for OWASP ASI04 Agentic Supply Chain Vulnerabilities.
-
Source Partly signed
github.com/bytedance/UI-TARS, Apache-2.0, last pushed 2026-01-27. 19% of the last 100 commits carry a verified signature; the rest can't be tied to a verified identity.
-
Review and release Weak
OpenSSF Scorecard rates the repository's practices 2.5/10 (scanned Oct 6, 2026). The checks that decide who can get a change released:
- Code Review1
- Branch Protection0
- Signed Releasesn/a
- Dangerous Workflow10
- Token Permissions0
- Pinned Dependencies0
All 18 Scorecard checks
Binary-Artifacts 10Branch-Protection 0CI-Tests 0CII-Best-Practices 0Code-Review 1Contributors 10Dangerous-Workflow 10Dependency-Update-Tool 0Fuzzing 0License 10Maintained 0Packaging -1Pinned-Dependencies 0SAST 0Security-Policy 0Signed-Releases -1Token-Permissions 0Vulnerabilities 0 -
Published packages Not linked
- PyPI ui-tars Source link isn't a GitHub repo No build attestation
Without a build attestation, nothing proves the published files were built from this repository.
-
What you install today No known advisories
Latest release checked: ui-tars 0.5.1. No published advisory affects it (OSV, checked 2026-10-09).
Changes to this chain
No change to package provenance, package source links, Scorecard coverage or license in HVTracker's daily snapshots of UI-TARS.
Evidence behind this score
Coverage A: 4 of 5 independent evidence types found.
- GitHub repository data
- Package downloads
- Supply-chain checks
- Public actions (missing)
- Community mentions
Verify this score yourself
Every build re-issues UI-TARS’s score as an Ed25519-signed credential, valid for 7 days. You can check it offline against HVTracker’s published key; if anyone changes a number after signing, verification fails.
What this score doesn’t check
It covers who ships the code, not what the code does. It doesn’t read tool descriptions or prompts for injected instructions, watch runtime behaviour, or find bugs nobody has disclosed yet. For that, run a content scanner before you connect it, such as Cisco MCP Scanner or Snyk Agent Scan.
How UI-TARS compares in Browser & Computer Use
- #35 OmniParser 44.9 +3.9
- #36 MolmoWeb 41.5 +0.5
- #37 UI-TARS 41.0 this agent
- #38 Understudy 35.0 −6.0
- #39 Agent-E 34.8 −6.2
Bars show each HVTrust score; the tick marks UI-TARS’s 41.0.
Where the 41.0 comes from
HVTrust dimensions vs the Browser & Computer Use average
41.0 / 100 · 100.0% confidenceUI-TARS Browser & Computer Use average (45 agents)
Quick Trust Read
How to read this: HVTrust (0–100) weighs supply-chain signals (provenance, OSSF Scorecard, signed commits, open license) alongside real-world adoption. Grade D reflects the trust score band: A ≥ 80, B ≥ 65, C ≥ 50, D < 50. Evidence coverage A is separate — it grades how many independent signal types back the score (4 of 5), so a high score on thin evidence stays visible. Full methodology →
Rank Trend
Activity & Reach
Analysis
Activity Inputs
38.1 / 100Common questions about UI-TARS
Does UI-TARS publish package provenance?
Does UI-TARS have an OpenSSF Scorecard?
Is UI-TARS actively maintained?
What license does UI-TARS use?
Are UI-TARS's commits signed?
Not a safety endorsement. HVTracker describes what public signals show, not whether a project is safe for your use case. Run your own security review before adopting in production.
AI agent surface
MCP, providers, tool surface
These runtime-trust fields — detected from public repo docs and manifests — contribute a bounded adjustment to this project's HVTrust score alongside supply-chain evidence. The exact values each field can add or subtract are documented in the methodology → Compare this surface across every listed agent in the capability matrix →
- MCP signal live
- External deps live
- Tool / plugin surface live
- Package provenance drift live
Detected changes to UI-TARS's runtime surface and supply-chain posture, from daily public-signal snapshots. A change here means our detectors see something different — a genuinely changed capability, or better evidence of an existing one.
Maintain UI-TARS?
For maintainers
HVTrust scores UI-TARS from public signals only — we never contact maintainers first. If a signal is wrong, stale, or missing (provenance you publish, a Scorecard you run, signed releases), tell us and we'll review it. Corrections are public and tracked on GitHub.
Reputation Timeline
Signal history
Embed Badge Badge guide for maintainers →
For maintainers
[](https://hvtracker.net/agents/ui-tars)
<a href="https://hvtracker.net/agents/ui-tars"><img src="https://hvtracker.net/badge/ui-tars.svg" alt="HVTrust"></a>
Other agents in Browser & Computer Use
GitHub REST API (repo, commits, stars, forks, license) · PyPI / pypistats (downloads, provenance) · OpenSSF Scorecard CLI · Algolia HN Search API
Each agent's signals refresh once daily across 6 staggered batches. Methodology v4.4 · Raw JSON