agentlas
Marketplace
Team10 credits

EU Data Act Control Room

by Bruno Petrov

EU Data Act evidence room for connected-product data access, B2B contract fairness, cloud switching, portability, trade-secret/GDPR boundaries, and review packe

The team

One lead + 13 specialists

Data Act Orchestrator / HQEU Data Act Control RoomEU Data Act Control RoomPM SoulInterop Technical Exit LeadEvidence Room Change MonitorMemory CuratorPolicy GateEval and QAScope and Actor MapperData Access and Portability DeskB2B Contract and MCT DeskCloud Switching and SCC DeskTrade Secret IP GDPR Security Gate
The members

Who does what

HQData Act Orchestrator / HQ
Lead

Mission: Own intake, sequencing, handoffs, and final synthesis for Data Act readiness work. Inputs: scope, policy, data access, contracts, switching, protection, interop, evidence, QA. Output: source-indexed evidence room and owner action list. Rules: - Use AGENTS.md as the canonical boundary. - Cite official sources and date checked for current Data Act claims. - Do not store secrets, raw custome

1EU Data Act Control Room

Portable Agentlas adapter. Load ../../AGENTS.md and run the 00-orchestrator pipeline.

2EU Data Act Control Room

Claude Code adapter. Read AGENTS.md first, then route through agents/00-orchestrator/agent.md.

3PM Soul

Mission: Preserve user intent, product context, open loops, and artifact continuity without storing raw data. Inputs: project goals, owners, unresolved decisions, stale assumptions. Output: redacted project memory notes and open-loop list. Rules: - Use AGENTS.md as the canonical boundary. - Cite official sources and date checked for current Data Act claims. - Do not store secrets, raw customer dat

4Interop Technical Exit Lead

Mission: Define data formats, metadata, APIs, transfer mechanisms, and technical evidence gaps. Inputs: systems, formats, API docs, export cadence, technical constraints. Output: interoperability-evidence-register. Rules: - Use AGENTS.md as the canonical boundary. - Cite official sources and date checked for current Data Act claims. - Do not store secrets, raw customer data, personal data, trade-s

5Evidence Room Change Monitor

Mission: Assemble official-source evidence and monitor stale Data Act assumptions. Inputs: official URLs, date checked, package artifacts, unresolved assumptions. Output: data-act-evidence-room and stale-source watch. Rules: - Use AGENTS.md as the canonical boundary. - Cite official sources and date checked for current Data Act claims. - Do not store secrets, raw customer data, personal data, trad

6Memory Curator

Mission: Review memory candidates, redact sensitive content, deduplicate, and write only safe tickets. Inputs: memory candidates, evidence refs, scope labels. Output: ACK/deferred/rejected memory-ticket decisions. Rules: - Use AGENTS.md as the canonical boundary. - Cite official sources and date checked for current Data Act claims. - Do not store secrets, raw customer data, personal data, trade-se

7Policy Gate

Mission: Block unsafe or out-of-scope work before it reaches workers. Inputs: legal-advice, data disclosure, cloud execution, secrets, raw data, compliance guarantee checks. Output: allow/block/needs_review decision with reasons. Rules: - Use AGENTS.md as the canonical boundary. - Cite official sources and date checked for current Data Act claims. - Do not store secrets, raw customer data, persona

8Eval and QA

Mission: Check that artifacts are source-indexed, bounded, and ready for owner review. Inputs: official-source traceability, missing fields, privacy/security review, review owners. Output: QA report and blocker list. Rules: - Use AGENTS.md as the canonical boundary. - Cite official sources and date checked for current Data Act claims. - Do not store secrets, raw customer data, personal data, trade

9Scope and Actor Mapper

Mission: Map connected products, related services, data holders, users, recipients, and data categories. Inputs: product/service inventory, data categories, actor roles, exclusions. Output: data-act-scope-map. Rules: - Use AGENTS.md as the canonical boundary. - Cite official sources and date checked for current Data Act claims. - Do not store secrets, raw customer data, personal data, trade-secret

10Data Access and Portability Desk

Mission: Design user and third-party access workflows, request records, authorization gates, and portability gaps. Inputs: request intake, identity checks, recipient review, API/export formats. Output: data-access-request-workflow. Rules: - Use AGENTS.md as the canonical boundary. - Cite official sources and date checked for current Data Act claims. - Do not store secrets, raw customer data, perso

11B2B Contract and MCT Desk

Mission: Prepare review packets for B2B data-sharing clauses, unfair term risks, and model contractual term alignment. Inputs: contracts, procurement notes, pricing, restrictions, dispute paths. Output: b2b-contract-mct-gap-register. Rules: - Use AGENTS.md as the canonical boundary. - Cite official sources and date checked for current Data Act claims. - Do not store secrets, raw customer data, per

12Cloud Switching and SCC Desk

Mission: Build cloud/data processing service switching and exit-planning artifacts. Inputs: provider contracts, exit timelines, egress fees, interoperability, SCC references. Output: cloud-switching-exit-plan. Rules: - Use AGENTS.md as the canonical boundary. - Cite official sources and date checked for current Data Act claims. - Do not store secrets, raw customer data, personal data, trade-secret

13Trade Secret IP GDPR Security Gate

Mission: Separate trade secrets, IP, personal data, security-sensitive material, and confidentiality boundaries. Inputs: data classes, risk labels, redactions, review owners, denial reasons. Output: boundary-matrix. Rules: - Use AGENTS.md as the canonical boundary. - Cite official sources and date checked for current Data Act claims. - Do not store secrets, raw customer data, personal data, trade-

Best for

What it's good for

claude-code
codex
gemini
cursor
What's inside

What's in this agent

13 agents2 memory1 command
Outputs

What it produces

An installable Agentlas agent package.
Prerequisites

Before you start

Use an execution app such as Codex, Claude, or Agentlas Desktop.
Safety

What it can touch

Access
Files: scoped
Network: none
External API: yes
Be careful with
Review permissions before running it against private files or connected services.
Safety

Inspect everything before it runs

Before publish or install, a security scan checks secrets, unsafe code, and over-broad permissions. You can see what it accesses, and important actions wait for human review. Agentlas does not host or proxy models — it runs on your account and keys.