Skip to content

[REVIEW] nist-csf-assessment: add evidence confidence and tier calibration #91

@stevesagronegocios673-ux

Description

Skill Being Reviewed

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

False Positive Analysis

Benign evidence that can be over-scored as mature:

Subcategory: GV.RM-02
Claimed current score: 3
Evidence: risk appetite mentioned in a board slide
Missing: approved risk appetite statement, review date, owner, communication record, exception handling, and evidence that the statement is used in ERM decisions

Why this is a false positive risk:
The skill already warns that CSF tiers apply at the organizational level, but Step 4 maps subcategory scores directly to tier labels and the output profile only has a free-form Evidence column. A roadmap slide, vendor dashboard, or policy draft can be treated as enough evidence for a high score unless the report captures artifact type, owner, scope, freshness, implementation coverage, and runtime/decision evidence.

Coverage Gaps

Missed variant 1: Tier labels used as subcategory maturity proof

ID.AM-01 current score: 3 / Tier 3
Evidence: asset inventory spreadsheet
Missing: update cadence, coverage percentage, authoritative source, lifecycle ownership, reconciliation signal

Why it should be caught:
CSF tiers describe organization-wide risk management integration, not individual subcategory compliance. Subcategory scores can be useful, but they should not be presented as literal tier determinations without separate organizational-tier evidence.

Missed variant 2: Current/target profiles lack evidence confidence

PR.AA-05 current score: 4
Evidence: IAM dashboard screenshot
Missing: apps/resources in scope, entitlement review sample, exception list, enforcement logs, last-tested date

Why it should be caught:
The primary CSF output is the Current and Target Organizational Profiles. Those profiles should include evidence confidence and Not Evaluable reason codes so unsupported claims do not silently become mature scores.

Missed variant 3: Govern and supply-chain outcomes lack decision evidence

GV.SC-07 score: 3
Evidence: vendor risk questionnaire completed
Missing: supplier criticality tier, contract requirement mapping, risk treatment decision, monitoring cadence, issue follow-up

Why it should be caught:
CSF 2.0 elevated GOVERN and supply-chain risk management. A questionnaire alone does not prove that supplier risk is recorded, prioritized, responded to, and monitored through the relationship.

Edge Cases

  • An organization can have Tier 3 governance but low maturity in a specific function, or strong technical controls with Tier 1 governance.
  • A target profile can intentionally stay below score 4 for low-risk areas; not every gap is a defect.
  • Vendor dashboards can support evidence, but they need scope, coverage, freshness, exception, and owner fields.
  • Community Profiles can be useful baselines but should not override the organization-specific mission, risk appetite, and regulatory context.

Remediation Quality

  • Fix resolves the vulnerability
  • Fix does not introduce new security issues
  • Fix does not break functionality
  • Issues found: Add evidence confidence levels, Not Evaluable reason codes, a normalized profile evidence table, and a separate organizational-tier evidence section. Clarify that 0-4 subcategory scoring is an assessment scale aligned to profile gaps, not a per-subcategory tier claim.

Comparison to Other Tools

Tool / Framework Catches this? Notes
NIST CSF 2.0 Partial Defines outcomes, Profiles, and Tiers, but assessment reports must preserve evidence quality.
NIST CSF 2.0 Reference Tool Partial Helps with taxonomy and examples, but does not validate organization evidence.
NIST SP 800-53 / CIS mappings Partial Helpful implementation references, but mappings alone do not prove CSF profile maturity.

Overall Assessment

Strengths:

  • Strong coverage of CSF 2.0 functions, categories, and subcategories.
  • Good explanation of GOVERN and profile development.
  • Useful roadmap output for current/target profile gaps.

Needs improvement:

  • Separate organizational-tier evidence from subcategory scoring.
  • Add profile evidence confidence and Not Evaluable reason codes.
  • Require owner, scope, freshness, implementation coverage, runtime/decision evidence, and exception tracking for profile entries.
  • Add pitfalls for vendor-dashboard evidence, roadmap-only maturity, and treating every target below 4 as a defect.

Sources Checked

Bounty Info

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