Loading ToolBench

Reading the current corpus…

ops0-cli MCP server · D · ToolBench

ops0-cli

Model Context Protocol server that exposes ops0 governance tools to AI agents (Claude Code, etc.) for policy enforcement and IaC compliance checking

2026-07-28+STDIO4 tools

57/100

Definition Quality68

Protocol Readiness45

Supportability45

Spec posture

Inferred effective spec 2026-07-28+·current baseline 2026-07-28

Static source inference · medium confidence

Current-spec patterns detected

Summary

Four tools with complete input schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. All tools are READ_ONLY and well-named with action verbs (list_, check_, get_). Descriptions are adequate (100-200 chars) but lack dependency hints and recovery guidance. Output structures not documented. Error handling deferred ('Implementation deferred' in check_compliance). STDIO transport caps overall potential.

Tools (4)

check_complianceread onlyauthsource verified68/100

Evaluate ops0 Rego policies against a Terraform plan JSON. Returns violations (or empty list if compliant).

get_finding_detailsread onlyauthsource verified68/100

Look up a discovery security check by templateId (e.g. aws-s3-public-access-block-disabled). Returns description + remediation.

get_iac_standardsread onlyauthsource verified68/100

Return the authoritative IaC output contract for the current ops0 org (required tags, provider-specific field names for tags/labels, README + docs structure, Mermaid syntax rules, project-scoped recommendations). Call this ONCE per session, before writing the first module - the response supersedes the static baseline in CLAUDE.md when the two disagree.

list_policiesread onlyauthsource verified67/100

List ops0 policies that apply to the current repository. Call this BEFORE generating Infrastructure-as-Code so the rules are in context.

Issues (5)

CRITICALtool-description ↗

No parameter descriptions in input schemas. 'planJsonPath' and 'templateId' lack inline descriptions explaining format, constraints, or examples. Rubric requires every parameter to have a description.

check_complianceget_finding_details

HIGHresponse-shaper ↗

Output schemas not documented. Tool handlers return 'content' blocks with 'text' type, but no formal schema definition visible. LLMs cannot plan downstream calls without knowing return structure.

list_policiescheck_complianceget_finding_detailsget_iac_standards

CRITICALrecovery-guide ↗

check_compliance implementation deferred ('Implementation deferred' in code). Tool is registered but non-functional. Error handling and recovery guidance missing.

check_compliance

MEDIUMtool-description ↗

No dependency hints in descriptions. get_iac_standards says 'Call this ONCE per session, before writing the first module' but list_policies lacks guidance on when to call it relative to other tools.

Recommendations

Add inline descriptions to all parameters in inputSchema. For 'planJsonPath', add: 'description': 'Filesystem path to Terraform plan JSON (output of terraform show -json plan.out). Must be readable by the server process.'
Document output schemas for each tool. Define the structure of 'content' array items, field types, and required fields. Example: 'Returns array of {type: string, text: string} objects.'
Complete check_compliance implementation. Currently deferred, implement OPA evaluation logic and return structured violation list with rule IDs, severity, and remediation hints.
Add error recovery guidance. When planJsonPath is invalid, return: 'File not found at <path>. Ensure terraform show -json plan.out has been run and the path is correct.' instead of generic error.
Migrate from STDIO to HTTP+SSE or Streamable HTTP transport to enable remote client access and improve protocol readiness score.

Score history

Overall score trend

First recorded score · v2 rubric

57/100

Scored	Grade	Overall	Spec posture	Rubric
2026-09-23	D	57	2026-07-28+	v2