RCA-MCP

[Adapted from Salesforce PyRCA — BSD-3-Clause] Identify anomalous metrics contributing to a Service Level Indicator (SLI) anomaly by comparing metric distributions in normal vs. incident windows. Uses z-score thresholding: metrics with |z| > epsilon in the incident window relative to the normal baseline are flagged as root cause candidates. Best used as a FIRST STEP in RCA to narrow down candidate metrics before applying more compute-intensive causal methods like rca_pyrca_ht_diagnosis or rca_pyrca_random_walk -- this one needs no causal graph at all, just two data windows. Requires the pyrca feature (Starter+). Args: params (EpsilonDiagnosisInput): - normal_data: baseline {metric: [values]} (min 3 per metric) - anomaly_data: incident window {metric: [values]} - sli_metric: the observed anomaly metric - epsilon: z-score threshold (default 3.0 = 3σ) Returns: str: JSON with root_causes (anomalous metrics ranked by |z_score|), all_metrics, sli_z_score, epsilon_threshold Attribution: Adapted from PyRCA EpsilonDiagnosis (Salesforce, BSD-3-Clause) Zhen et al. (2022) ε-Diagnosis

[Adapted from Salesforce PyRCA — BSD-3-Clause] Hypothesis-testing RCA with descendant adjustment (HT-ADJ / CIRCA). Tests whether the SLI anomaly can be statistically explained by causal propagation from each ancestor node. Applies descendant adjustment to reduce indirect cause scores and surface true root causes. This is the most statistically rigorous PyRCA algorithm and is recommended when you have a well-validated causal graph and sufficient pre-anomaly data -- reach for rca_pyrca_epsilon_diagnosis instead if you don't have a graph yet, or rca_pyrca_random_walk if you have a graph but not enough pre-anomaly history for a hypothesis test. Requires the pyrca feature (Starter+). Args: params (HTDiagnosisInput): - data: {metric: [values]} full time series - adjacency: causal graph - sli_metric: observed anomaly metric - anomaly_start_idx: index where anomaly starts - significance: p-value threshold (default 0.05) - use_descendant_adjustment: enable HT-ADJ (default True) Returns: str: JSON with root_causes (is_root_cause=true), all_results, method (HT or HT-ADJ) Attribution: Adapted from PyRCA HT/CIRCA concept (Salesforce, BSD-3-Clause) Shen et al. (2022) CIRCA; Zheng et al. (2023) arXiv:2306.11417

[Adapted from Salesforce PyRCA — BSD-3-Clause] Graph-based root cause localisation via personalised PageRank random walk. Propagates backward through a causal adjacency graph from the SLI node, weighting transitions by anomaly scores to compute root cause probabilities. Requires the pyrca feature (Starter+) and, unlike rca_pyrca_epsilon_diagnosis, needs an adjacency graph you already have (from rca_graph_get's "adjacency" format, or hand-built) plus precomputed anomaly scores per metric -- it doesn't compute those scores itself. Args: params (RandomWalkInput): - adjacency: {source: {target: weight}} causal graph - anomaly_scores: {metric: score} anomaly magnitudes - sli_metric: starting node - restart_prob: personalisation (higher = proximity-weighted) Returns: str: JSON with root_causes ranked by composite_score, converged, iterations Attribution: Adapted from PyRCA random walk concept (Salesforce, BSD-3-Clause)

Validate the PyRCA integration setup and report which strategy is active — a read-only diagnostic, not an action. Takes no parameters beyond authentication (token/client_id); there is nothing else to configure on this call. Run this once before your first rca_pyrca_epsilon_diagnosis, rca_pyrca_ht_diagnosis, or rca_pyrca_random_walk call if you're unsure which strategy is active, or if a PyRCA call errors unexpectedly — the response's "recommendations" field will say what to fix. Checks: - Strategy B (pure Python): always available, no extra setup - Strategy A (subprocess): requires sfr-pyrca in .venv_pyrca - sklearn version in host env vs PyRCA's requirement - Attribution compliance (BSD-3-Clause notice present) Returns: str: JSON with strategy_active, sklearn_version, sfr_pyrca_available, compliance, recommendations

Generate a comparative report across 2–10 RCA results, showing consensus root causes, model agreement percentages, and per-model summaries. Use this instead of rca_analysis_compare when you want a shareable formatted document rather than raw comparison JSON. `format: "html"` requires Starter+ (Free plan gets a plan_required error and should use the default "markdown" instead). Args: params (ReportCompareInput): - result_ids: 2–10 result IDs to compare (from rca_analysis_run or rca_analysis_list_results) - format: "markdown" (default, all plans) or "html" (Starter+) - title: report title, up to 200 chars - save: persist the report server-side for later retrieval (default true) Returns: str: Comparative report (text/html) with consensus_root_causes table

Generate a styled, professional report from a SINGLE RCA analysis result -- use rca_report_compare instead if you want a report covering multiple results together. Supported formats (plan-gated -- Free: markdown only, Starter+: adds html/pdf, Pro+: adds excel; requesting an ungated format returns a security_violation error, not a silent downgrade): pdf — Professional PDF with tables, score bars, and styled sections (requires reportlab; falls back to plaintext if not installed) html — Styled HTML with CSS — embeddable in dashboards or emails (requires jinja2; falls back to minimal HTML) excel — 4-sheet Excel workbook: Summary, Root Causes, Actions, Metadata (requires openpyxl) markdown — Plain Markdown; always available; good for GitHub/Slack/Notion All formats include: - Executive summary with metric cards - Root cause ranking table with priority badges - Recommended actions (IMMEDIATE / MONITOR / TRACK) - Analysis metadata - Optional raw output appendix Args: params (ReportGenerateInput): - result_id: source RCA result - format: pdf | html | excel | markdown - title: custom report title - include_raw: include model output appendix - save: persist report to storage Returns: str: JSON with content_b64 (bytes formats), content_text (text formats), byte_size, format, report_id (if saved), storage_path (if saved)