Reality Graph Verification Tools

Measured conclusion: Deep-tested. L2 outcome: tools_ok.

Read-only AI coding tools for change verification, release readiness, capacity, and guidance.

Add to your agent

Copy this into your MCP client config:

{
  "mcpServers": {
    "Reality Graph Verification Tools": {
      "url": "https://realitygraph.dev/api/mcp",
      "transport": "streamable-http"
    }
  }
}

Measured tools (10)

• check_verification_debt read-only

  Estimate a software team's verification debt from team parameters. Computes the four published metrics (generation-to-verification ratio, review depth, unverified-merge rate, two-week churn) and an annual cost estimate, with the full calculation path, labeled assumptions, thresholds, and sources (GitClear, Sonar, Faros, Veracode). Deterministic arithmetic from published models - no benchmark claims. Only team_size is required; every additional parameter refines the estimate. Set lang='de' for a German report.

  Parameters:

  • team_size (required) — integer — Number of developers on the team (required)
  • lang — string — Report language (default: en)
  • prs_per_month — integer — Total merged PRs per month (default: team_size x prs_per_engineer_per_month)
  • hourly_rate_eur — number — Loaded cost per engineer hour in EUR (default: 75, assumption)
  • ai_share_percent — number — Share of merges that are AI-assisted, in percent (default: 60, assumption)
  • ai_merges_per_month — integer — AI-assisted merges per month (enables the unverified-merge rate)
  • merged_loc_per_week — number — Merged changed lines of code per week (enables the GVR and review-depth metrics)
  • rework_rate_percent — number — Share of AI-assisted changes reworked for a defect within 14 days, in percent (default: 2, the illustrative rate from /cost-of-verification-debt - replace it with your own reason-coded rate)
  • two_week_churn_percent — number — Share of new lines revised or reverted within 14 days, in percent. A warning signal in the metrics block; it never enters the cost model, because it measures lines and the cost model counts changes
  • reviewer_hours_per_week — number — Reviewer hours actually spent per week (enables the GVR metric)
  • hours_per_reworked_change — number — Average hours per reworked change (default: 6, assumption)
  • prs_per_engineer_per_month — number — Merged PRs per engineer per month (default: 20, derived from the published worked report on /measure-verification-debt)
  • incident_allowance_eur_per_year — number — Annual incident allowance in EUR (default: 0; add one only when you have a locally defined incident class, frequency and expected-loss method)
  • ai_merges_with_evidence_per_month — integer — AI-assisted merges per month with recorded validation evidence (enables the unverified-merge rate)
  • review_reconstruction_hours_per_pr — number — Average reviewer hours spent reconstructing intent per AI-assisted PR (default: 0.5, assumption)
  • substantive_review_comments_per_week — number — Substantive review comments per week, excluding bots and nitpicks (enables the review-depth metric)

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

• lint_task_spec read-only

  Check whether a free-text work order for an AI coding agent is verifiable BEFORE handing it over. Heuristic, deterministic lint of the task's form against the four building blocks of a checkable task (goal, boundaries, acceptance criteria, validation plan) plus rule checks (vague adjectives without numbers, unnamed unhappy paths, missing file anchors). Returns a status table with evidence, the concrete questions that close each gap, and a fill-in skeleton. It checks form, not content — no LLM, nothing stored. Set lang='de' for a German report.

  Parameters:

  • task (required) — string — The work order / task text you intend to give an AI coding agent (English or German)
  • lang — string — Report language (default: en)

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

• get_task_contract_template read-only

  Returns Reality Graph's free fill-in template (v0) for a verifiable task contract: goal, non-goals, boundaries (may change / must not change / forbidden), 3-7 yes/no acceptance criteria, validation plan, expected evidence, assumptions, open questions — with a filled example and fill-in guidance. Write the contract before an AI agent runs; verify the result against it after. format='json' returns a machine-fillable JSON structure; default is a compact markdown skeleton. Set lang='de' for German. Static content, nothing stored.

  Parameters:

  • lang — string — Language (default: en)
  • format — string — Template format (default: markdown)

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

• validate_task_contract read-only

  Deterministically validates a FILLED task contract (the JSON structure from get_task_contract_template): completeness of goal/non-goals/boundaries, decidability of each acceptance criterion (vague words, missing measurable markers), automated checks in the validation plan, expected evidence, and leftover placeholders. Returns a verdict (PASS / PASS WITH WARNINGS / FAIL), four dimension scores, and a concrete fix per finding. Validates form and completeness, not correctness. No LLM, nothing stored. lang='de' for German.

  Parameters:

  • contract (required) — string — The filled task contract as a JSON string (structure from get_task_contract_template, format='json')
  • lang — string — Report language (default: en)

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

• get_verification_report_template read-only

  Returns the free fill-in template (v0) for a verification report — the artifact you write right after an AI-assisted run: task recap, files changed AND files confirmed untouched, validation results per acceptance criterion (not authored by the generating model), what was skipped, limitations, and the explicit decision. format='json' for a machine-fillable structure; default is a compact markdown file. Static content, nothing stored. lang='de' for German.

  Parameters:

  • lang — string — Language (default: en)
  • format — string — Template format (default: markdown)

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

• search read-only

  Full-text search over the Reality Graph knowledge base on AI coding verification: 40+ glossary definitions, 700+ FAQ answers, sourced statistics, and article summaries on verification debt, AI code review, spec-vs-implementation checking, EU compliance (EU AI Act, GDPR, NIS2), and AI coding governance — in English and German. Returns matching documents with title, URL, and snippet. Use fetch to read a result.

  Parameters:

  • query (required) — string — Search query (English or German)
  • lang — string — Restrict results to one language (default: both)

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

• fetch read-only

  Fetch a document from the Reality Graph knowledge base by id (as returned by search, e.g. '/verification-debt') or by full realitygraph.dev URL. Returns the document's summary, definitions, key facts, FAQ, and sources as text, plus the canonical URL.

  Parameters:

  • id (required) — string — Document id from search results, or a realitygraph.dev URL

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

• plan_change_verification read-only

  Turn explicit change characteristics into a risk tier, required automated checks, manual scenarios, evidence, release blockers, role handoff, and canonical Reality Graph guidance. Use before implementation or review. It does not inspect code and never invents a confidence score.

  Parameters:

  • change_summary (required) — string — Plain-language summary of the change
  • change_types (required) — array — Technical and risk-relevant change types
  • blast_radius (required) — string — Largest expected impact boundary
  • rollback (required) — string — Current rollback or recovery state
  • lang — string — Response language (default: en)

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

• check_release_readiness read-only

  Return GO, CONDITIONAL, or NO_GO from supplied acceptance-criterion results, check evidence, rollback, monitoring, limitations, and independent review. The verdict is deliberately based only on supplied evidence; this tool does not inspect code, CI, or a deployment.

  Parameters:

  • change_summary (required) — string — Plain-language summary of the change
  • change_types (required) — array — Technical and risk-relevant change types
  • blast_radius (required) — string — Largest expected impact boundary
  • rollback (required) — string — Current rollback or recovery state
  • acceptance_criteria_passed (required) — integer
  • acceptance_criteria_failed (required) — integer
  • acceptance_criteria_not_run (required) — integer
  • checks (required) — array
  • rollback_ready (required) — boolean
  • monitoring_ready (required) — boolean
  • known_limitations_recorded (required) — boolean
  • independent_review (required) — boolean
  • lang — string — Response language (default: en)

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

• calculate_verification_capacity read-only

  Calculate weekly review demand, utilization, capacity gap, supported change throughput, and changes lacking evidence from measured team inputs. No cost model, benchmark, or hidden industry assumption is applied; the output shows the arithmetic and a concrete balancing action.

  Parameters:

  • ai_changes_per_week (required) — integer
  • average_review_minutes_per_change (required) — number
  • available_reviewer_hours_per_week (required) — number
  • evidence_coverage_percent (required) — number
  • lang — string — Response language (default: en)
  • two_week_churn_percent — number

  Hints: read-only yes · destructive no · idempotent not declared · open-world no

Publisher website

Measured by MCPYP probing. Untested fields are omitted, never estimated. Listing is always free; MCPYP never proxies calls or stores user tokens.