security

DE//AGENT currently lists 8 resources under security: 4 tutorials, 2 skills, 1 hook. Everything below is derived from the live registry — counts, capabilities, verification state and security posture come from the curated records, never from marketing copy.

Also present: 1 template.

Documented learning objectives across these resources:

• Read a tool schema and predict its side effects — via MCP Tool Contract Auditor
• Classify tools by blast radius — via MCP Tool Contract Auditor
• Write a least-privilege MCP configuration — via MCP Tool Contract Auditor
• Classify columns into PII categories — via PII Classifier
• Choose masking vs hashing vs tokenisation — via PII Classifier
• Record a data-classification tag — via PII Classifier
• Separate agent read/recommend from write/execute — via Keep an agent read-only until a human approves
• Encode the safety policy as hooks — via Keep an agent read-only until a human approves

7 more objectives live on the individual resource pages.

These resources map to stages: Agentic Data Engineering (5), Reliability (2), Cloud (1). Follow the four-phase roadmap to place this topic inside a progression instead of learning it in isolation.

Every entry, its shape and its trust state — sorted by the registry's composite score. This table is the honest version of a marketing comparison: no sponsored placements, no unverifiable claims.

How scores are computed: source quality score.

Highest composite score first:

1. MCP Tool Contract Auditor score 40.5 — advanced
2. PII Classifier score 34.4 — intermediate
3. Keep an agent read-only until a human approves score 34.15 — beginner

Scores rank provenance, completeness and trust — read how verification works before adopting anything.

Start here if you're new: Keep an agent read-only until a human approves, Secret Scan on Commit.

Reach for these with production experience: MCP Tool Contract Auditor, Production gotchas — untrusted content, destructive actions, permission scope, Terraform AWS Data Platform.

Security-flagged entries: MCP Tool Contract Auditor (1 flag), PII Classifier (3 flags), Secret Scan on Commit (2 flags), Terraform AWS Data Platform (5 flags). Flags mean the record declares capabilities like credential use or write access — read the security posture on each page before connecting.

• MCP Tool Contract Auditor — Inspects an MCP server's tool list and schemas, flags tools that can write, delete, spend money or exfiltrate data, and produces a least-privilege connection recommendation. details
• PII Classifier — Scans column names and sample values for PII categories, then proposes classification tags and a masking/pseudonymisation policy. details
• Keep an agent read-only until a human approves — Coding agents can do real data-engineering work, but can also ship a schema change or drop a table. Walk the Agent-in-the-Loop Review pattern: separate read/recommend from write/ex… details
• Audit an MCP server's tool contracts — Before connecting an MCP server, audit its tools for blast radius. Read each tool schema, classify every tool as Read, Write, Destructive, Spend or Exfiltrate, and produce a least-… details

Real install commands from the registry — read the source before running any of them:

npx de-agent-platform add mcp-tool-contract-auditor

MCP Tool Contract Auditor — review first

npx de-agent-platform add pii-classifier

PII Classifier — review first

amp template new de-agent/terraform-aws-data-platform ./infra

Terraform AWS Data Platform — review first

Adjacent topics: tutorial (4 shared), review (3 shared), agentic (3 shared), least-privilege (3 shared), Python (2 shared), mcp (2 shared).

Study path: the roadmap places this topic in a progression · the scoring model explains the numbers · the trust rule catalog shows every security check these entries passed through.