Skip to main content
Glama
Sign In
mcp-server-scf

mcp-server-scf

by MarkAC007
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Remote

scf_assessment_read

Read-only

Read AI evidence assessments to retrieve file verdicts, summaries, review queues, window assessments, and control composites. Choose a method to inspect assessment status, scores, findings, and costs.

Instructions

Read AI evidence assessments: a file's verdict, the file summary, review queues, window assessments (list, one, versions, summary), and control composites. Pick with method.

Input Schema

NameRequiredDescriptionDefault
tierNofile (default): entries carry file_id, act with scf_review_evidence_assessment. window: entries carry window_assessment_id, act with scf_review_window_assessment_verdict Used by: review_queue.
limitNoPage size (1–200, default 50) [review_queue] | Page size (1–100, default 10) [list_windows] | Page size (1–500, default 100) [list_control_composites] Used by: review_queue, list_windows, list_control_composites.
cursorNoOpaque pagination cursor — pass next_cursor from a prior response Used by: list_control_composites.
domainNoFilter by SCF domain code (e.g., 'BCD', 'GOV', 'AST') Used by: list_control_composites.
methodYesfile_assessment: Get the AI assessment for an evidence file: status, relevance score (0–100), findings, summary, audit metadata (model, tokens, cost). Findings name catalogue-mapped controls, not only in-scope ones. file_summary: Get aggregate AI assessment metrics for the organization dashboard: total assessed, counts by status, unassessed count, average relevance score, and total cost in cents. review_queue: List AI verdicts awaiting a human decision, worst first (read — viewer role). tier=file (default) lists per-file verdicts; tier=window is the web app's Awaiting-confirmation queue. list_windows: List recent windowed AI assessments for an evidence item (newest first). Each entry includes window bounds, frequency, file IDs, coverage, status, relevance score, findings, and cost. window: Get one windowed AI assessment by ID. Returns full detail: window bounds, frequency, file IDs, coverage, expected artifact types, status, relevance score, findings, summary, hashes, tokens, cost. window_versions: List every AI verdict a window assessment has received, newest first (read — viewer role). Each version is frozen as reached (model, prompt version, designations) plus any human decision on it. window_summary: Get aggregate windowed-assessment metrics for the organization dashboard: total windows assessed, counts by status (including `insufficient_sample`), average relevance score, and total cost in cents. control_composite: Get the rolled-up assessment composite for one SCF control: composite score, status band, included/missing evidence IDs, mandatory gaps, per-window detail. 404 if no composite row exists yet (async). list_control_composites: List rolled-up assessment composites for the org. Cursor-paginated, worst-band first (insufficient → sufficient). Filter by status/domain/computation_version. Pass next_cursor to page forward.
offsetNoPagination offset (default 0) [review_queue] | Pagination offset — number of results to skip (default 0) [list_windows] Used by: review_queue, list_windows.
org_idYesOrganization UUID — obtain from scf_list_organizations
scf_idNoSCF control identifier in DOMAIN-NN format (e.g., 'AST-01', 'GOV-02') Used by: control_composite.
statusNoQueue filter (default awaiting) [review_queue] | Comma-separated composite_status values to include (e.g., 'insufficient,partial'). Valid values: insufficient, insufficient_sample, partial, pending, no_evidence, sufficient [list_control_composites] Used by: review_queue, list_control_composites.
file_idNoEvidence file UUID — obtain from scf_list_evidence_files Used by: file_assessment.
evidence_idNoEvidence ID (e.g., 'ERL-IAM-001') — obtain from scf_list_evidence [file_assessment] | Evidence ID (e.g., 'E-IAM-01') — obtain from scf_list_evidence [list_windows] Used by: file_assessment, list_windows.
assessment_idNoWindowed assessment UUID — obtain from scf_list_window_assessments [window] | Window assessment UUID — obtain from scf_list_window_assessments [window_versions] Used by: window, window_versions.
computation_versionNoRestrict to composites computed at this algorithm version Used by: list_control_composites.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Addedv4.1.0

TDQS

B3.4/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

The description is consistent with readOnlyHint=true and adds scope by listing the assessment types covered. However, it does not add behavioral context such as role requirements, async composite availability, or pagination behavior beyond what annotations and schema already provide.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is a single front-loaded sentence with no wasted wording, and it efficiently lists the supported assessment categories. The ending "Pick with method" is terse but functional, though the sentence could better structure the method-selection instruction.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 9-method, 13-parameter polymorphic read tool, the description is minimal and relies heavily on the detailed method enum in the schema. It gives enough orientation to select the tool, but it does not explain return shapes or method-specific behavior at the top level.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so parameter meanings are already fully documented in the input schema. The description adds only the high-level instruction to select behavior with method, which is useful but already reflected in the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states a specific read operation for AI evidence assessments and enumerates the main assessment categories: file verdict, file summary, review queues, window assessments, and control composites. It is clearly distinct from a write tool, though it does not explicitly name sibling alternatives.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The phrase "Pick with method" gives essential dispatch guidance for a polymorphic tool, but it does not explain when to use this tool versus related read tools or when each method is appropriate. The detailed method guidance lives in the schema rather than in the description.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools