---
schema: 1
type: video
title: 'We Vetted 2000 AI Skills Before They Reached Developers — Lucas Palma, Nubank'
description: >-
  An AI skill is a piece of code you hand a model to extend what it can do, and
  once engineers start sharing skills with each other, each one becomes a supply
  chain risk, more so inside a regulated bank. Lucas Palma's security team at
  Nubank built Skill Vector to sit between a skill and the internal marketplace,
  so nothing reaches developers unvetted. Every skill is scanned first with
  deterministic checks, for destructive shell commands and the like, then with
  an LLM for the context those checks miss, and only then does it get a decision
  and permissions scoped to who will use it. Running that gate over more than
  2,000 skills surfaced real problems, since a single skill can carry many, and
  fed them into the bank's vulnerability management program with approval gates
  and human confirmation. What worked was pairing deterministic scans with LLM
  review; what needed work was the guidance the system gave and the habit of
  running skills locally before vetting. The lesson he leaves is simple: treat
  skills like any other dependency, and only what clears the gate belongs in the
  marketplace. Speaker info: - https://www.linkedin.com/in/lucaspalma/
  Timestamps: 0:00 - Introduction: making code safe at a bank 1:32 - AI skills
  as a supply chain risk 2:50 - What counts as an AI skill 3:57 - The extra
  weight of a regulated environment 6:07 - From plugins to a vetted marketplace
  6:58 - What Skill Vector does 7:37 - Deterministic checks, then the LLM 10:00
  - Scanning over two thousand skills 11:22 - What worked and what needed
  improvement 13:30 - Approval gates and human confirmation 14:23 - Next steps
  and policies
canonical: >-
  https://liquidgold.cc/channel/aidotengineer/explore/video/we-vetted-2000-ai-skills-before-they-reached-develo-c21041c90225
markdown: >-
  https://liquidgold.cc/channel/aidotengineer/video/we-vetted-2000-ai-skills-before-they-reached-develo-c21041c90225.md
llms: null
parent: 'https://liquidgold.cc/channel/aidotengineer.md'
channel: aidotengineer
updated: '2026-10-24T06:29:44.144Z'
language: null
children: []
videoId: iKQ78wyJEXU
durationSeconds: 984
publishedAt: '2026-07-29T22:00:06.000Z'
searchChunks:
  - ordinal: 0
    text: null
    startSeconds: 95
    endSeconds: 105
    watch: 'https://liquidgold.cc/watch/iKQ78wyJEXU?t=95'
    heldReason: rights
  - ordinal: 1
    text: null
    startSeconds: 105
    endSeconds: 177
    watch: 'https://liquidgold.cc/watch/iKQ78wyJEXU?t=105'
    heldReason: rights
  - ordinal: 2
    text: null
    startSeconds: 189
    endSeconds: 244
    watch: 'https://liquidgold.cc/watch/iKQ78wyJEXU?t=189'
    heldReason: rights
  - ordinal: 3
    text: null
    startSeconds: 244
    endSeconds: 257
    watch: 'https://liquidgold.cc/watch/iKQ78wyJEXU?t=244'
    heldReason: rights
  - ordinal: 4
    text: null
    startSeconds: 257
    endSeconds: 374
    watch: 'https://liquidgold.cc/watch/iKQ78wyJEXU?t=257'
    heldReason: rights
  - ordinal: 5
    text: null
    startSeconds: 257
    endSeconds: 374
    watch: 'https://liquidgold.cc/watch/iKQ78wyJEXU?t=257'
    heldReason: rights
  - ordinal: 6
    text: null
    startSeconds: 384
    endSeconds: 418
    watch: 'https://liquidgold.cc/watch/iKQ78wyJEXU?t=384'
    heldReason: rights
  - ordinal: 7
    text: null
    startSeconds: 430
    endSeconds: 464
    watch: 'https://liquidgold.cc/watch/iKQ78wyJEXU?t=430'
    heldReason: rights
  - ordinal: 8
    text: null
    startSeconds: 464
    endSeconds: 476
    watch: 'https://liquidgold.cc/watch/iKQ78wyJEXU?t=464'
    heldReason: rights
  - ordinal: 9
    text: null
    startSeconds: 476
    endSeconds: 608
    watch: 'https://liquidgold.cc/watch/iKQ78wyJEXU?t=476'
    heldReason: rights
  - ordinal: 10
    text: null
    startSeconds: 476
    endSeconds: 608
    watch: 'https://liquidgold.cc/watch/iKQ78wyJEXU?t=476'
    heldReason: rights
  - ordinal: 11
    text: null
    startSeconds: 608
    endSeconds: 621
    watch: 'https://liquidgold.cc/watch/iKQ78wyJEXU?t=608'
    heldReason: rights
  - ordinal: 12
    text: null
    startSeconds: 688
    endSeconds: 699
    watch: 'https://liquidgold.cc/watch/iKQ78wyJEXU?t=688'
    heldReason: rights
  - ordinal: 13
    text: null
    startSeconds: 699
    endSeconds: 814
    watch: 'https://liquidgold.cc/watch/iKQ78wyJEXU?t=699'
    heldReason: rights
  - ordinal: 14
    text: null
    startSeconds: 814
    endSeconds: 827
    watch: 'https://liquidgold.cc/watch/iKQ78wyJEXU?t=814'
    heldReason: rights
  - ordinal: 15
    text: null
    startSeconds: 827
    endSeconds: 872
    watch: 'https://liquidgold.cc/watch/iKQ78wyJEXU?t=827'
    heldReason: rights
  - ordinal: 16
    text: null
    startSeconds: 883
    endSeconds: 984
    watch: 'https://liquidgold.cc/watch/iKQ78wyJEXU?t=883'
    heldReason: rights
---


# We Vetted 2000 AI Skills Before They Reached Developers — Lucas Palma, Nubank

An AI skill is a piece of code you hand a model to extend what it can do, and once engineers start sharing skills with each other, each one becomes a supply chain risk, more so inside a regulated bank. Lucas Palma's security team at Nubank built Skill Vector to sit between a skill and the internal marketplace, so nothing reaches developers unvetted. Every skill is scanned first with deterministic checks, for destructive shell commands and the like, then with an LLM for the context those checks miss, and only then does it get a decision and permissions scoped to who will use it. Running that gate over more than 2,000 skills surfaced real problems, since a single skill can carry many, and fed them into the bank's vulnerability management program with approval gates and human confirmation. What worked was pairing deterministic scans with LLM review; what needed work was the guidance the system gave and the habit of running skills locally before vetting. The lesson he leaves is simple: treat skills like any other dependency, and only what clears the gate belongs in the marketplace. Speaker info: - https://www.linkedin.com/in/lucaspalma/ Timestamps: 0:00 - Introduction: making code safe at a bank 1:32 - AI skills as a supply chain risk 2:50 - What counts as an AI skill 3:57 - The extra weight of a regulated environment 6:07 - From plugins to a vetted marketplace 6:58 - What Skill Vector does 7:37 - Deterministic checks, then the LLM 10:00 - Scanning over two thousand skills 11:22 - What worked and what needed improvement 13:30 - Approval gates and human confirmation 14:23 - Next steps and policies

Canonical: https://liquidgold.cc/channel/aidotengineer/explore/video/we-vetted-2000-ai-skills-before-they-reached-develo-c21041c90225 · Channel: [AI Engineer](https://liquidgold.cc/channel/aidotengineer.md)

> Search chunk text is held until public-source provenance is verified.

## Document chunks

> Missing: no stored document chunks are available.
