---
type: "post"
id: "we-ve-integrated-tencenthunyuan-s-ai-infra-guard-367d80"
slug: "we-ve-integrated-tencenthunyuan-s-ai-infra-guard-367d80"
title: "We’ve integrated @TencentHunyuan's AI-Infra-Guard (AIG) into ClawScan, the open-source command-line tool that powers security review on ClawHub."
description: "Thesis: Adding a second independent scanner can broaden the evidence available to a platform’s security review, but benchmark results should not be read as a guarantee for live uploads."
retrieval_nugget: "Thesis: Adding a second independent scanner can broaden the evidence available to a platform’s security review, but benchmark results should not be read as a guarantee for live uploads."
published_at: "2026-10-03"
updated_at: "2026-10-03"
record_date: "2026-10-03"
date_kind: "published_at"
topics: []
source_urls: ["https://openclaw.ai/blog/tencent-aig-joins-clawscan"]
schema_version: "newruntime-agent-readable-v0.2"
stable_id: "post:we-ve-integrated-tencenthunyuan-s-ai-infra-guard-367d80"
status: "published"
visuals: []
editorial_provenance: {"schema_version":"newruntime-editorial-copy-v1","content_status":"source_grounded_final","final_copy_sha256":"sha256:8732273854241641f4f880cb11f68e17c33325fc5d3ea542c8eb79c8bde28609","reviewed_at":"2026-10-03T02:01:33.076Z","source_evidence_count":1,"verified_claim_count":3,"site_analysis_schema_version":"newruntime-site-analysis-v1","site_object_kind":"field_note","observed_fact_count":3,"implication_count":1,"watch_condition_count":1,"related_record_count":0}
analysis: {"schema_version":"newruntime-site-analysis-v1","object_kind":"field_note","thesis":"Adding a second independent scanner can broaden the evidence available to a platform’s security review, but benchmark results should not be read as a guarantee for live uploads.","observed_facts":[{"text":"OpenClaw says every skill and plugin uploaded to ClawHub now runs through Tencent’s AI-Infra-Guard (AIG) as part of review.","source_urls":["https://openclaw.ai/blog/tencent-aig-joins-clawscan"]},{"text":"ClawScan runs AIG and NVIDIA’s SkillSpector independently, then has an AI judge consider both findings alongside the uploaded files.","source_urls":["https://openclaw.ai/blog/tencent-aig-joins-clawscan"]},{"text":"OpenClaw reports 86.9% agreement with benchmark labels and correct classification of 98.6% of malicious cases on a fixed 556-case subset of SkillTrustBench.","source_urls":["https://openclaw.ai/blog/tencent-aig-joins-clawscan"]}],"mechanism":"The integration preserves distinct scanner assessments for a later review step rather than treating one scanner as the sole verdict.","why_now":"AIG is now included in ClawHub’s review path, and the announcement describes a feedback loop in which anonymized disagreements and confirmed false positives are shared with Tencent for regression tests and scanner improvements.","implications":["The integration makes cross-scanner disagreement an input to review and iteration, rather than hiding it behind a single scanner score."],"evidence_boundary":"These are figures reported by OpenClaw for a specific benchmark subset; they do not show the rate of missed threats or false positives across all live ClawHub uploads.","watch_conditions":["Watch for later reporting that measures false positives, missed threats, or changes in benchmark results after the feedback loop updates the scanners."],"related_records":[],"new_branch_reason":"No prior New Runtime record is supplied here, so this is a distinct analytical record about this integration and its evidence limits, not a continuation of an assumed earlier item."}
editor: {"role":"editor","name":"Andrey Reshetnikov","profile":"https://newruntime.com/owner-profile.md","profile_json":"https://newruntime.com/owner-profile.json","same_as":["https://www.linkedin.com/in/reshetnikov1/","https://t.me/lapetuse"]}
routes: {"html":"https://newruntime.com/posts/we-ve-integrated-tencenthunyuan-s-ai-infra-guard-367d80/","markdown":"https://newruntime.com/posts/we-ve-integrated-tencenthunyuan-s-ai-infra-guard-367d80.md","json":"https://newruntime.com/posts/we-ve-integrated-tencenthunyuan-s-ai-infra-guard-367d80.json"}
---

# We’ve integrated @TencentHunyuan's AI-Infra-Guard (AIG) into ClawScan, the open-source command-line tool that powers security review on ClawHub.

## Retrieval answer

Thesis: Adding a second independent scanner can broaden the evidence available to a platform’s security review, but benchmark results should not be read as a guarantee for live uploads.

Thesis: Adding a second independent scanner can broaden the evidence available to a platform’s security review, but benchmark results should not be read as a guarantee for live uploads. OpenClaw says every skill and plugin uploaded to ClawHub now runs through Tencent’s AI-Infra-Guard (AIG) as part of review. It also says ClawScan runs AIG and NVIDIA’s SkillSpector independently, then has an AI judge consider both findings alongside the uploaded files.

Mechanism: The integration preserves distinct scanner assessments for a later review step rather than treating one scanner as the sole verdict. OpenClaw reports that AIG and SkillSpector surfaced different risks even when using the same AI model. Why now: AIG is now included in ClawHub’s review path, and the announcement describes a feedback loop in which anonymized disagreements and confirmed false positives are shared with Tencent for regression tests and scanner improvements. In a fixed 556-case subset of SkillTrustBench, OpenClaw reports 86.9% agreement with benchmark labels and correct classification of 98.6% of malicious cases.

Evidence boundary: These are figures reported by OpenClaw for a specific benchmark subset; they do not show the rate of missed threats or false positives across all live ClawHub uploads. First implication: The integration makes cross-scanner disagreement an input to review and iteration, rather than hiding it behind a single scanner score. First watch condition: Watch for later reporting that measures false positives, missed threats, or changes in benchmark results after the feedback loop updates the scanners. No prior New Runtime record is supplied here, so this is a distinct analytical record about this integration and its evidence limits, not a continuation of an assumed earlier item.

Editor: [Andrey Reshetnikov](https://newruntime.com/owner-profile.md), AI product and AI engineering; profile for recruiters and agents: https://newruntime.com/owner-profile.json

## Continue in the graph

Next records for an agent, linked to their Markdown twins. Replace `.md` with `.json` for structured fields, including `next_reads`.

- collection: [Field Notes](https://newruntime.com/posts.json): Open the latest editorial analysis.
- collection: [Source Ledger](https://newruntime.com/source-ledger.json): Inspect the public source evidence graph.
- whole graph: [discovery graph](https://newruntime.com/discovery-graph.json) (typed nodes and edges) and [site index](https://newruntime.com/site-index.md)