Skip to content

[REVIEW] nist-csf-assessment: add supplier concentration and exit evidence gates #1166

@99INFLUENCERS

Description

@99INFLUENCERS

Skill Being Reviewed

Skill name: nist-csf-assessment
Skill path: skills/compliance/nist-csf-assessment/

False Positive Analysis

Benign-looking scenario that can be over-scored as mature supply-chain risk management:

assessment_scope: B2B SaaS platform
current_evidence:
  GV.OC-05:
    external_dependencies:
      - primary_cloud_provider
      - email_delivery_provider
      - payment_processor
      - customer_support_saas
  GV.SC-04:
    supplier_inventory:
      - name: primary_cloud_provider
        criticality: critical
      - name: payment_processor
        criticality: high
  GV.SC-05:
    supplier_contracts:
      - security_addendum: present
      - breach_notice_clause: present
      - confidentiality_clause: present
  GV.SC-07:
    annual_vendor_review: completed
claimed_result:
  GV.SC: "3 - Repeatable"

Why this is a false positive:

The current skill asks whether suppliers are inventoried, prioritized, included in contracts, assessed before acquisition, monitored during the relationship, and handled after the conclusion of a partnership. Those prompts are directionally correct, but they can still let an assessor score supply-chain risk as mature when the only evidence is an inventory plus contract clauses.

That misses whether the organization has evidence for concentration risk, fourth-party dependency visibility, supplier exit/offboarding controls, alternate-provider feasibility, and operational resilience if a critical supplier fails or is terminated. A supplier can be known and contractually governed while still creating an untested single point of failure.

Coverage Gaps

Missed variant 1: critical supplier concentration and substitutability are not evaluated

critical_services:
  authentication:
    supplier: identity_saas_a
    backup_supplier: none
    tested_failover: false
  payments:
    supplier: payment_processor_a
    backup_supplier: "contract signed, not integrated"
    tested_failover: false
  email_delivery:
    supplier: email_saas_a
    backup_supplier: email_saas_b
    tested_failover: false

Why it should be caught:

GV.OC-05 and GV.SC-04 identify dependencies and supplier criticality, but the skill does not require concentration-risk fields such as sole-source dependency, viable substitute, switching time, tested failover, contractual portability, or residual business impact. Without those fields, a high-quality supplier inventory can hide that recovery depends on a supplier that cannot be replaced in an outage or contract dispute.

Missed variant 2: fourth-party and subprocessor dependencies are not mapped to critical services

Primary supplier: customer_support_saas
Critical service: support ticket intake and customer PII handling

Known contract evidence:
- DPA present
- security addendum present

Missing fourth-party evidence:
- hosting cloud provider
- AI summarization provider
- observability/logging provider
- outsourced support subcontractor
- data residency for each subprocessor
- notice and approval path for subprocessor changes

Why it should be caught:

GV.SC-07 and GV.SC-09 require understanding and monitoring supplier risk over the relationship and product/service life cycle. For SaaS and managed services, material cyber risk often sits in fourth parties and subprocessors. The skill should require a dependency chain evidence table so supplier criticality is not evaluated only at the direct vendor layer.

Missed variant 3: supplier exit/offboarding is mentioned but not evidenced

Supplier termination checklist:
- contract ended
- final invoice paid

Missing technical evidence:
- SSO app disabled
- SCIM/API tokens revoked
- shared Slack/Teams channels removed
- vendor VPN account disabled
- webhook secrets rotated
- customer data export verified
- data deletion certificate received
- DNS/CNAME/vendor-hosted subdomain removed
- backups and support attachments covered by deletion/retention terms

Why it should be caught:

GV.SC-10 says C-SCRM plans should include provisions after a partnership or service agreement ends, but the current process only asks the question at a high level. It should require exit evidence for identity, network access, data return/deletion, DNS/vendor-hosted assets, secrets, integrations, and retained backup/support data. Otherwise a terminated supplier can retain access or data while the assessment marks GV.SC-10 as covered.

Missed variant 4: supplier incident participation is not tested

Contract says supplier will notify incidents within 72 hours.

Missing operational proof:
- named incident contact and escalation path
- evidence package expectations
- joint tabletop or notification drill
- supplier status page / API health dependency in incident runbooks
- recovery-time dependencies for customer-facing services

Why it should be caught:

GV.SC-08 requires relevant suppliers and third parties to be included in incident planning, response, and recovery. A contract clause is weaker evidence than an exercised escalation path or tabletop. The skill should separate contractual incident clauses from tested operational participation.

Edge Cases

  • Cloud providers can be both critical suppliers and infrastructure platforms; the report should capture shared-responsibility evidence, support plan dependency, region/service concentration, and backup-region limits.
  • A supplier may be low spend but high criticality, such as DNS, identity, code-signing, package registry, payment routing, or transactional email.
  • Vendor marketplaces and app integrations can create suppliers that bypass procurement, so SSO/OAuth/SaaS discovery evidence matters.
  • Open-source or package ecosystem dependencies may not have contracts, but they can still be critical third-party dependencies that need owner, substitute, and monitoring evidence.
  • Supplier exit evidence often lives across procurement, legal, IAM, IT, DNS, and application teams; the skill should allow not_evaluable_owner_unavailable rather than silently lowering or guessing the score.

Remediation Quality

  • Fix resolves the vulnerability
  • Fix doesn't introduce new security issues
  • Fix doesn't break functionality
  • Issues found: Add explicit C-SCRM evidence gates to the skill so an inventory-plus-contract review cannot over-score supply-chain maturity. The improvement should add concentration, fourth-party, exit/offboarding, and supplier-incident-test evidence fields to GV.OC-05, GV.SC-04, GV.SC-07, GV.SC-08, GV.SC-09, and GV.SC-10.

Recommended additions:

  1. Add a Supplier Concentration and Substitutability Matrix with supplier, service dependency, criticality, sole-source status, replacement option, switching time, tested failover, contractual portability, and residual impact.
  2. Add a Fourth-Party / Subprocessor Chain table with direct supplier, fourth party, service/data handled, region, change-notice mechanism, evidence source, and monitoring owner.
  3. Add a Supplier Exit Evidence checklist covering identity revocation, network access, API tokens, webhooks, DNS/vendor-hosted assets, data export, data deletion/retention, backups, and support artifacts.
  4. Add supplier incident drill fields: named contacts, escalation SLA, joint tabletop date, evidence package expected, recovery dependency, and not-evaluable reason.
  5. Add not-evaluable reason codes for missing supplier owner, missing fourth-party list, missing exit evidence, missing failover test, and missing supplier incident contact.

Comparison to Other Tools

Tool / Source Catches this? Notes
NIST CSF 2.0 GV.SC Partial Defines C-SCRM outcomes, including monitoring across the life cycle and provisions after relationship end, but the local skill needs evidence fields to operationalize those outcomes.
NIST SP 1305 C-SCRM Quick Start Guide Partial Focuses on using the CSF GV.SC category to establish and operate C-SCRM capability; the skill should translate that into auditable review fields.
NIST SP 800-161 Rev. 1 Partial Provides deeper C-SCRM practices across the organization and system life cycle, but the current skill does not capture concentration, fourth-party, or exit evidence explicitly.
Generic vendor-risk questionnaires Partial Often ask for contracts, attestations, and questionnaires, but frequently miss technical offboarding and tested substitutability unless customized.

Overall Assessment

Strengths:

  • Strong CSF 2.0 structure and useful GOVERN coverage.
  • Correctly includes GV.SC-01 through GV.SC-10 and identifies supplier inventory, contracts, due diligence, monitoring, incident planning, and relationship conclusion as assessment topics.
  • Good guardrails against fabricated CSF IDs and CSF 1.1 / 2.0 terminology drift.

Needs improvement:

  • Supplier criticality should include concentration risk and substitutability, not only inventory and priority.
  • Fourth-party and subprocessor dependencies need explicit evidence mapping for SaaS, cloud, support, monitoring, AI, and managed-service suppliers.
  • GV.SC-10 needs technical offboarding evidence, not just a high-level partnership-conclusion question.
  • Supplier incident readiness should distinguish contract clauses from exercised escalation and recovery participation.

Priority recommendations:

  1. Add supplier concentration and tested substitutability fields to GV.OC-05 and GV.SC-04.
  2. Add fourth-party/subprocessor chain evidence to GV.SC-07 and GV.SC-09.
  3. Add supplier exit/offboarding evidence gates to GV.SC-10.
  4. Add supplier incident drill evidence to GV.SC-08.
  5. Add not_evaluable_* reason codes for supplier evidence controlled by procurement, legal, IAM, or vendor owners.

Sources Checked

This review is distinct from #91/#93 because it focuses on supply-chain concentration, fourth-party, exit, and supplier incident evidence, not CSF Tier/profile scoring or evidence confidence generally. It is distinct from #712 because it focuses on C-SCRM operational dependency evidence, not privacy/civil-liberties impact.

Bounty Info

  • I have read and agree to the CONTRIBUTING.md bounty terms
  • Preferred payment method: PayPal can be coordinated privately after acceptance.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions