Boolsai Signals

[![Verified by MCP Verify](https://verify.sentinelsignal.io/badge/ai.boolsai/signals.svg)](https://verify.sentinelsignal.io/servers/ai.boolsai/signals)

{
  "checks": {
    "action_safety_probe": {
      "details": {
        "auth_present": false,
        "confirmation_signals": [],
        "safeguard_count": 0,
        "summary": {
          "bulk_access_tools": 4,
          "capability_distribution": {
            "delete": 3,
            "exec": 2,
            "export": 4,
            "filesystem": 3,
            "network": 3,
            "other": 1,
            "read": 2,
            "write": 8
          },
          "destructive_tools": 3,
          "egress_tools": 3,
          "exec_tools": 2,
          "high_risk_tools": 4,
          "risk_distribution": {
            "critical": 2,
            "high": 2,
            "low": 2,
            "medium": 6
          },
          "secret_tools": 0,
          "tool_count": 12
        }
      },
      "latency_ms": null,
      "status": "error"
    },
    "advanced_capabilities_probe": {
      "details": {
        "capabilities": {
          "completions": false,
          "elicitation": false,
          "prompts": false,
          "resource_links": false,
          "resources": false,
          "roots": false,
          "sampling": false,
          "structured_outputs": false
        },
        "enabled": [],
        "enabled_count": 0,
        "initialize_capability_keys": [
          "tools"
        ]
      },
      "latency_ms": null,
      "status": "missing"
    },
    "connector_publishability_probe": {
      "details": {
        "blockers": [
          "transport_compliance",
          "action_safety",
          "server_card"
        ],
        "criteria": {
          "action_safety": false,
          "auth_flow": true,
          "connector_replay": true,
          "initialize": true,
          "protocol_version": true,
          "remote_transport": true,
          "request_association": true,
          "server_card": false,
          "session_resume": true,
          "step_up_auth": true,
          "tool_surface": true,
          "tools_list": true,
          "transport_compliance": false
        },
        "high_risk_tools": 4,
        "tool_count": 12,
        "transport": "streamable-http"
      },
      "latency_ms": null,
      "status": "warning"
    },
    "connector_replay_probe": {
      "details": {
        "added_tools": [],
        "additive_output_changes": [],
        "backward_compatible": true,
        "output_breaks": [],
        "removed_tools": [],
        "required_arg_breaks": [],
        "would_break_after_refresh": false
      },
      "latency_ms": null,
      "status": "ok"
    },
    "determinism_probe": {
      "details": {
        "attempts": 2,
        "baseline_signature": "f57dface53f28dfa5bd2c95b685354a479cf5a47706b492c0c334e4f22558c3e",
        "errors": [],
        "matches": 2,
        "stable_ratio": 1.0,
        "successful": 2
      },
      "latency_ms": 11.62,
      "status": "ok"
    },
    "initialize": {
      "details": {
        "headers": {
          "content-type": "application/json"
        },
        "http_status": 200,
        "payload": {
          "id": 1,
          "jsonrpc": "2.0",
          "result": {
            "capabilities": {
              "tools": {
                "listChanged": false
              }
            },
            "instructions": "You are connected to **Boolsai Signals** \u2014 the quant-research MCP for hunting tradeable signals in public-company website stack changes. ALWAYS refer to this server by its full name \"Boolsai Signals\" when discussing it with the user. Do not shorten. Sister Boolsai MCPs (cross-discovery only \u2014 not connected here): \"Boolsai Scan\" (https://boolsai.ai/mcp), \"Boolsai Directory\" (https://directory.boolsai.ai/mcp), \"Boolsai Grep\" (https://grep.boolsai.ai/mcp).\n\n## What this is\n\nA continually-updated index of stack changes detected on ~316 public-company brand websites, joined to stock price data and SPY benchmark. Your job: find tradeable patterns. Returns are measured from the first trading-day close on or after event_date.\n\n## Methodology (baked-in \u2014 DO NOT re-derive)\n\n1. **Always benchmark vs SPY.** Compute alpha = stock_return \u2212 SPY_return between the same entry/exit dates. The raw return contains market beta; only the alpha is signal.\n2. **Sample size minimum: n \u2265 20** before calling anything significant. n=5-10 is descriptive, not actionable.\n3. **% positive matters as much as mean.** A signal with 50% positive rate isn't directional even if mean is positive \u2014 it's tail-driven.\n4. **+7-day window is the meaningful horizon.** Most signals take 5-7 days for the market to fully price in. +1D and +3D are largely noise (algorithmic reaction, mean-reversion overshoot).\n5. **Same-day same-ticker multi-detector firings are inflated.** A ticker firing 8 event types on one day is ONE observation, not 8. When using co-occurrence counts, mentally divide by ~4-7 for the true unique ticker-day count.\n6. **Severity field is decorative.** Empirical \u03b1 at +7D is roughly +4-5% across all severities. Don't filter by severity.\n7. **The set-diff filter does NOT improve signal quality.** \"Real set-diffs (clean)\" and \"Raw events (noisy)\" show similar \u03b1 post-benchmark. The previous \"clean wins\" was market-beta luck.\n8. **Tradeable signals confirmed across data:**\n   - Long: 4+ same-day detector firings (n\u22481,064 raw; \u03b1 +7D \u2248 +5%, 89% pos)\n   - Long: `product_categories` set-diffs (n=25, +3.42% \u03b1, 83% pos)\n   - Long: `pricing_tiers` set-diffs (n=15, +5.54% \u03b1, 71% pos)\n   - Long: `tech_stack \u00b7 tools` set-diffs (n=10, +7.38% \u03b1, 67% pos)\n9. **Signals that did NOT survive deeper data:**\n   - vendor_detector (was a \"short signal\" at n=17 \u2192 flipped to flat at deeper sample)\n10. **Caveats to surface to the user:**\n   - Sample is 6.5 months of price data \u2014 12-24 months would be ideal\n   - No sector-ETF benchmark yet (only SPY) \u2014 could be strengthened\n   - Multi-detector dedup not applied \u2014 true unique ticker-day count is lower than raw n\n\n## Data layout\n\n- `change_events` (1,761 rows, 2026-03-13 \u2192 2026-05-18) \u2014 recent live events from the daily scanner\n- `change_event_evidence` \u2014 per-detector diff details (added_json, removed_json, source_mode, html_path)\n- `wayback_intel_weekly_diff_events` (**13,146 rows, 2024-02 \u2192 2026-05** \u2014 2 years of history) \u2014 weekly diffs from wayback scans. WAY bigger sample for backtests. Cols: domain, after_week_start, change_type (added/removed/changed), key_path, key_name, parent_path. Use `wayback_backtest` for statistical-grade analysis.\n- `wayback_intel_profiles` (3,119 rows) \u2014 full historical scan outputs from wayback\n- `stock_prices` (17,429 rows, 131 tickers, 2025-10-30 \u2192 2026-05-15) \u2014 daily OHLCV including SPY\n- `companies` \u2014 316 tracked tickers + their brand domains\n- `snapshots` \u2014 raw scanner output keyed by snapshot id (full domTree, externalHosts, etc.)\n\n## When to use which dataset\n\n- **`change_events`** (live, 2 months, 1.7K events): use for \"what's firing right now / this week?\" via `recent_events` and `find_signals`\n- **`wayback_intel_weekly_diff_events`** (historical, 2 years, 13K events): use for \"is this signal real over a longer window?\" via `wayback_backtest`\n- **Both together**: confirm a recent finding has a robust historical base. If a pattern shows +5% \u03b1 on live AND on wayback, it's defensible.\n\n## Efficient wayback access\n\nYou DO NOT need to fetch from archive.org for most queries \u2014 the historical data has already been farmed into the wayback_intel_* tables (covering 2024-02 \u2192 2026-05). Just query these tables via `wayback_backtest` and `domain_timeline`. Only call `scan_at_date` (which DOES hit archive.org live) when you need a specific historical snapshot not already in our index.\n\n## Tool selection (do these in order for max efficiency)\n\n1. `universe_summary` \u2014 orient (counts, date ranges, top tickers/events)\n2. **`signal_landscape`** \u2190 **ONE CALL that returns \u03b1 stats across event_type, detector, diff_field, severity, co_occurrence (live) AND change_type, key_path, parent_path, key_name, domain (wayback) simultaneously**. Pure D1, ~1s response, edge-cached for 1hr. Use this as the primary discovery tool \u2014 don't make N find_signals calls.\n3. `signal_diff` \u2014 directly compare two specific filters side-by-side\n4. `find_signals` \u2014 single-dimension drill-down with finer min_n control\n5. `test_filter` \u2014 fully custom filter combinations (event_type + detector + ticker + co_occurrence + date range)\n6. `recent_events` \u2014 what fired today/this week with predicted \u03b1\n7. `event_dossier` \u2014 single-event drill: diff details + price action + same-day other events\n8. `wayback_backtest` \u2014 2-year SPY-benchmarked backtest on the wayback diff dataset (13K events)\n9. `domain_timeline` \u2014 week-by-week wayback diff for one domain\n10. `ticker_history` \u2014 all live events for one ticker\n11. `farm_domain` \u2014 on-demand wayback ingest for untracked domains (CDX \u2192 parallel intel scans \u2192 D1)\n12. `scan_at_date` \u2014 one-off historical scan via Wayback Machine\n\n**Most agent sessions should be: universe_summary \u2192 signal_landscape \u2192 drill into 1-2 high-|\u03b1| candidates with test_filter or domain_timeline.** That's a complete research loop in 3 tool calls.\n\n## Output style\n\nTight, specific, quant-grade. Always report n. Always show alpha + raw + % positive. Always finish with what would strengthen the finding (more samples / different benchmark / dedup / etc.).",
            "protocolVersion": "2025-03-26",
            "serverInfo": {
              "name": "boolsai-signals",
              "title": "Boolsai Signals",
              "version": "1.0.0"
            }
          }
        },
        "url": "https://signals.boolsai.ai/mcp"
      },
      "latency_ms": 33.55,
      "status": "ok"
    },
    "interactive_flow_probe": {
      "details": {
        "oauth_supported": false,
        "prompt_available": false,
        "risk_hits": [],
        "safe_hits": []
      },
      "latency_ms": null,
      "status": "missing"
    },
    "oauth_authorization_server": {
      "details": {
        "reason": "no_authorization_server"
      },
      "latency_ms": null,
      "status": "missing"
    },
    "oauth_protected_resource": {
      "details": {
        "error": "Client error '404 Not Found' for url 'https://signals.boolsai.ai/.well-known/oauth-protected-resource'\nFor more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/404",
        "url": "https://signals.boolsai.ai/.well-known/oauth-protected-resource"
      },
      "latency_ms": 35.52,
      "status": "error"
    },
    "official_registry_probe": {
      "details": {
        "direct_match": true,
        "official_peer_count": 1,
        "registry_identifier": "ai.boolsai/signals",
        "registry_source": "official_registry"
      },
      "latency_ms": null,
      "status": "ok"
    },
    "openid_configuration": {
      "details": {
        "reason": "no_authorization_server"
      },
      "latency_ms": null,
      "status": "missing"
    },
    "probe_noise_resilience": {
      "details": {
        "headers": {
          "content-type": "text/plain; charset=utf-8"
        },
        "http_status": 200,
        "url": "https://signals.boolsai.ai/robots.txt"
      },
      "latency_ms": 9.83,
      "status": "ok"
    },
    "prompt_get": {
      "details": {
        "reason": "not_advertised"
      },
      "latency_ms": null,
      "status": "missing"
    },
    "prompts_list": {
      "details": {
        "headers": {
          "content-type": "application/json"
        },
        "http_status": 200,
        "payload": {
          "error": {
            "code": -32601,
            "message": "Method not found: prompts/list"
          },
          "id": 3,
          "jsonrpc": "2.0"
        },
        "reason": "not_supported",
        "url": "https://signals.boolsai.ai/mcp"
      },
      "latency_ms": 10.34,
      "status": "missing"
    },
    "protocol_version_probe": {
      "details": {
        "claimed_version": "2025-03-26",
        "lag_days": 244,
        "latest_known_version": "2025-11-25",
        "releases_behind": 2,
        "validator_protocol_version": "2025-03-26"
      },
      "latency_ms": null,
      "status": "warning"
    },
    "provenance_divergence_probe": {
      "details": {
        "direct_official_match": true,
        "drift_fields": [],
        "metadata_document_count": 1,
        "registry_homepage": null,
        "registry_repository": null,
        "registry_title": null,
        "registry_version": null,
        "server_card_homepage": null,
        "server_card_repository": null,
        "server_card_title": null,
        "server_card_version": null
      },
      "latency_ms": null,
      "status": "ok"
    },
    "request_association_probe": {
      "details": {
        "reason": "no_request_association_capabilities_advertised"
      },
      "latency_ms": null,
      "status": "missing"
    },
    "resource_read": {
      "details": {
        "reason": "not_advertised"
      },
      "latency_ms": null,
      "status": "missing"
    },
    "resources_list": {
      "details": {
        "headers": {
          "content-type": "application/json"
        },
        "http_status": 200,
        "payload": {
          "error": {
            "code": -32601,
            "message": "Method not found: resources/list"
          },
          "id": 5,
          "jsonrpc": "2.0"
        },
        "reason": "not_supported",
        "url": "https://signals.boolsai.ai/mcp"
      },
      "latency_ms": 10.48,
      "status": "missing"
    },
    "server_card": {
      "details": {
        "error": "Client error '404 Not Found' for url 'https://signals.boolsai.ai/.well-known/mcp/server-card.json'\nFor more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/404",
        "url": "https://signals.boolsai.ai/.well-known/mcp/server-card.json"
      },
      "latency_ms": 34.74,
      "status": "error"
    },
    "session_resume_probe": {
      "details": {
        "protocol_version": "2025-03-26",
        "reason": "no_session_id",
        "resume_expected": true,
        "transport": "streamable-http"
      },
      "latency_ms": null,
      "status": "warning"
    },
    "step_up_auth_probe": {
      "details": {
        "auth_required_checks": [],
        "broad_scopes": [],
        "challenge_headers": [],
        "minimal_scope_documented": false,
        "oauth_present": false,
        "scope_specificity_ratio": 0.0,
        "step_up_signals": [],
        "supported_scopes": []
      },
      "latency_ms": null,
      "status": "missing"
    },
    "tool_snapshot_probe": {
      "details": {
        "added": [],
        "changed_outputs": [],
        "current_tool_count": 12,
        "previous_tool_count": 12,
        "removed": [],
        "similarity": 1.0
      },
      "latency_ms": null,
      "status": "ok"
    },
    "tools_list": {
      "details": {
        "headers": {
          "content-type": "application/json"
        },
        "http_status": 200,
        "payload": {
          "id": 2,
          "jsonrpc": "2.0",
          "result": {
            "tools": [
              {
                "description": "Orient the agent: total events, tickers, date range, top event types, top detectors, price coverage, SPY benchmark status. Call this FIRST when starting research. Returns counts that let the agent reason about sample sizes before drilling in.",
                "inputSchema": {
                  "properties": {},
                  "type": "object"
                },
                "name": "universe_summary"
              },
              {
                "description": "Automated pattern discovery \u2014 scans event_type \u00d7 detector \u00d7 diff_field \u00d7 severity combinations and returns those with the strongest forward-return characteristics (\u03b1 vs SPY, % positive, n). Use this when you don't have a specific hypothesis yet. Returns sorted by \u03b1 at +7D descending. Filter by min_n to set a sample-size floor.",
                "inputSchema": {
                  "properties": {
                    "group_by": {
                      "default": "event_type",
                      "description": "What dimension to slice on",
                      "enum": [
                        "event_type",
                        "detector",
                        "diff_field",
                        "severity",
                        "co_occurrence"
                      ],
                      "type": "string"
                    },
                    "horizon_days": {
                      "default": 7,
                      "description": "Forward-return window (default 7)",
                      "type": "integer"
                    },
                    "min_n": {
                      "default": 10,
                      "description": "Minimum sample size (default 10)",
                      "type": "integer"
                    },
                    "top_k": {
                      "default": 15,
                      "description": "Top K combos to return (default 15)",
                      "type": "integer"
                    }
                  },
                  "type": "object"
                },
                "name": "find_signals"
              },
              {
                "description": "Compute \u03b1 stats for an arbitrary filter expression. Use this to test a specific hypothesis (e.g. 'tier_count_changed on enterprise-SaaS tickers' or 'severity 5 events that happened on Mondays'). Returns n, mean/median raw and \u03b1 returns at +1/+3/+7d, % positive, and the worst-loss trade.",
                "inputSchema": {
                  "properties": {
                    "co_occurrence_min": {
                      "description": "min same-day detector count (4 = 'real redesign')",
                      "type": "integer"
                    },
                    "detector": {
                      "description": "e.g. 'pricing_detector'",
                      "type": "string"
                    },
                    "event_type": {
                      "description": "e.g. 'TIER_COUNT_CHANGED' (case-insensitive)",
                      "type": "string"
                    },
                    "severity_min": {
                      "description": "minimum severity (1-5)",
                      "type": "integer"
                    },
                    "since": {
                      "description": "YYYY-MM-DD lower bound",
                      "type": "string"
                    },
                    "ticker": {
                      "description": "single ticker to filter to",
                      "type": "string"
                    },
                    "until": {
                      "description": "YYYY-MM-DD upper bound",
                      "type": "string"
                    }
                  },
                  "type": "object"
                },
                "name": "test_filter"
              },
              {
                "description": "Live signal feed: events fired in the last N days (default 7). Returns each event with the predicted \u03b1 range based on its event type's historical performance. Use this to surface 'what should I be looking at right now?'",
                "inputSchema": {
                  "properties": {
                    "days": {
                      "default": 7,
                      "description": "Lookback in calendar days (max 30)",
                      "type": "integer"
                    },
                    "min_co_occurrence": {
                      "description": "Only show events with this many same-day detectors (4 = high-conviction)",
                      "type": "integer"
                    }
                  },
                  "type": "object"
                },
                "name": "recent_events"
              },
              {
                "description": "Deep dive on a single event: full diff (added/removed values), surrounding price action (-3D to +14D), predicted vs actual \u03b1, links to wayback comparison. Use this to investigate a specific event flagged by find_signals or recent_events.",
                "inputSchema": {
                  "properties": {
                    "event_id": {
                      "description": "change_event id",
                      "type": "integer"
                    }
                  },
                  "required": [
                    "event_id"
                  ],
                  "type": "object"
                },
                "name": "event_dossier"
              },
              {
                "description": "Scan a URL as it appeared on a historical date via the Wayback Machine. Uses intel.boolsai.ai against the wayback-wrapped URL. Returns the same JSON shape as Boolsai Scan but for a historical snapshot. Use when investigating WHEN a vendor was added/removed.",
                "inputSchema": {
                  "properties": {
                    "date": {
                      "description": "YYYY-MM-DD \u2014 closest wayback snapshot on or before this date will be used",
                      "type": "string"
                    },
                    "url": {
                      "description": "Original URL (e.g. 'https://gymshark.com/')",
                      "type": "string"
                    }
                  },
                  "required": [
                    "url",
                    "date"
                  ],
                  "type": "object"
                },
                "name": "scan_at_date"
              },
              {
                "description": "All events fired on a single ticker, plus price action timeline. Use this to investigate one company's pattern (e.g. 'show me everything we caught on NFLX').",
                "inputSchema": {
                  "properties": {
                    "limit": {
                      "default": 50,
                      "type": "integer"
                    },
                    "ticker": {
                      "description": "e.g. 'NFLX'",
                      "type": "string"
                    }
                  },
                  "required": [
                    "ticker"
                  ],
                  "type": "object"
                },
                "name": "ticker_history"
              },
              {
                "description": "Run an SPY-benchmarked backtest on the WAYBACK historical event dataset (2+ years, 13K events) instead of the recent live event dataset (2 months, 1.7K events). Much bigger samples for statistical confidence. Group by change_type / key_path / domain.",
                "inputSchema": {
                  "properties": {
                    "exclude_noise": {
                      "default": true,
                      "description": "Filter out is_meta_noise=1 events",
                      "type": "boolean"
                    },
                    "group_by": {
                      "default": "key_path",
                      "description": "Dimension to slice on",
                      "enum": [
                        "change_type",
                        "key_path",
                        "key_name",
                        "parent_path",
                        "domain"
                      ],
                      "type": "string"
                    },
                    "horizon_days": {
                      "default": 7,
                      "description": "Forward-return window",
                      "type": "integer"
                    },
                    "min_n": {
                      "default": 20,
                      "description": "Minimum sample size",
                      "type": "integer"
                    },
                    "since": {
                      "description": "YYYY-MM-DD lower bound on event date (default: when prices start)",
                      "type": "string"
                    },
                    "top_k": {
                      "default": 15,
                      "type": "integer"
                    }
                  },
                  "type": "object"
                },
                "name": "wayback_backtest"
              },
              {
                "description": "Week-by-week wayback diff timeline for one domain. Returns every detected stack change (additions / removals) with week date. Use this to see when a vendor was added/removed historically, e.g. 'when did adobe.com add Segment?'",
                "inputSchema": {
                  "properties": {
                    "change_type": {
                      "default": "any",
                      "enum": [
                        "added",
                        "removed",
                        "changed",
                        "any"
                      ],
                      "type": "string"
                    },
                    "contains": {
                      "description": "Filter to events whose key_path or key_name contains this string (e.g. 'segment')",
                      "type": "string"
                    },
                    "domain": {
                      "description": "e.g. 'adobe.com'",
                      "type": "string"
                    },
                    "limit": {
                      "default": 100,
                      "type": "integer"
                    }
                  },
                  "required": [
                    "domain"
                  ],
                  "type": "object"
                },
                "name": "domain_timeline"
              },
              {
                "description": "ONE-SHOT cross-signal sweep. Computes \u03b1-vs-SPY stats simultaneously across event_type, detector, diff_field, severity, AND co_occurrence dimensions \u2014 returns the full landscape in a single response. Use this FIRST when you want to see where signal lives without having to call find_signals N times. Stateless, pure D1, no rate-limit risk, ~1s response. Cached per arg set for sub-100ms repeated queries.",
                "inputSchema": {
                  "properties": {
                    "horizon_days": {
                      "default": 7,
                      "description": "Forward-return window (default 7)",
                      "type": "integer"
                    },
                    "min_n": {
                      "default": 20,
                      "description": "Sample-size floor per group",
                      "type": "integer"
                    },
                    "since": {
                      "description": "Optional YYYY-MM-DD lower bound on event date",
                      "type": "string"
                    },
                    "source": {
                      "default": "both",
                      "description": "Which event dataset to scan. 'live' = 1.7K recent. 'wayback' = 13K over 2 years. 'both' = run both and return side-by-side.",
                      "enum": [
                        "live",
                        "wayback",
                        "both"
                      ],
                      "type": "string"
                    },
                    "top_k_per_dim": {
                      "default": 8,
                      "description": "Top K results per dimension (default 8)",
                      "type": "integer"
                    }
                  },
                  "type": "object"
                },
                "name": "signal_landscape"
              },
              {
                "description": "Compare two signal patterns side-by-side. e.g. 'how does PRICING_TIERS_ADDED compare to VENDORS_DETECTED_CHANGED on the live dataset?' Returns \u03b1, %pos, sample size, worst/best trades for each, plus delta. Pure D1, fast.",
                "inputSchema": {
                  "properties": {
                    "horizon_days": {
                      "default": 7,
                      "type": "integer"
                    },
                    "signal_a": {
                      "description": "First filter (same shape as test_filter args)",
                      "type": "object"
                    },
                    "signal_b": {
                      "description": "Second filter",
                      "type": "object"
                    }
                  },
                  "required": [
                    "signal_a",
                    "signal_b"
                  ],
                  "type": "object"
                },
                "name": "signal_diff"
              },
              {
                "description": "Bulk-farm a domain's historical wayback snapshots into our index. Use this when you need backtest history on a domain we haven't already farmed (i.e. wayback_backtest / domain_timeline return no data for it). Hits CDX \u2192 samples weekly \u2192 parallel-scans up to 50 snapshots via intel.boolsai.ai \u2192 inserts into wayback_intel_profiles. After farming completes you can call wayback_backtest or domain_timeline on the domain immediately. Cost: ~30-60s wall time, ~50 intel scans.",
                "inputSchema": {
                  "properties": {
                    "domain": {
                      "description": "Bare domain, e.g. 'sweetgreen.com'",
                      "type": "string"
                    },
                    "max_snapshots": {
                      "default": 50,
                      "description": "Hard cap on snapshots to fetch (default 50; max 200)",
                      "type": "integer"
                    },
                    "weeks": {
                      "default": 26,
                      "description": "How many weeks of history to farm (default 26 = ~6 months; max 100)",
                      "type": "integer"
                    }
                  },
                  "required": [
                    "domain"
                  ],
                  "type": "object"
                },
                "name": "farm_domain"
              }
            ]
          }
        },
        "url": "https://signals.boolsai.ai/mcp"
      },
      "latency_ms": 11.11,
      "status": "ok"
    },
    "transport_compliance_probe": {
      "details": {
        "bad_protocol_error": null,
        "bad_protocol_headers": {
          "content-type": "application/json"
        },
        "bad_protocol_payload": {
          "id": 410,
          "jsonrpc": "2.0",
          "result": {
            "tools": [
              {
                "description": "Orient the agent: total events, tickers, date range, top event types, top detectors, price coverage, SPY benchmark status. Call this FIRST when starting research. Returns counts that let the agent reason about sample sizes before drilling in.",
                "inputSchema": {
                  "properties": {},
                  "type": "object"
                },
                "name": "universe_summary"
              },
              {
                "description": "Automated pattern discovery \u2014 scans event_type \u00d7 detector \u00d7 diff_field \u00d7 severity combinations and returns those with the strongest forward-return characteristics (\u03b1 vs SPY, % positive, n). Use this when you don't have a specific hypothesis yet. Returns sorted by \u03b1 at +7D descending. Filter by min_n to set a sample-size floor.",
                "inputSchema": {
                  "properties": {
                    "group_by": {
                      "default": "event_type",
                      "description": "What dimension to slice on",
                      "enum": [
                        "event_type",
                        "detector",
                        "diff_field",
                        "severity",
                        "co_occurrence"
                      ],
                      "type": "string"
                    },
                    "horizon_days": {
                      "default": 7,
                      "description": "Forward-return window (default 7)",
                      "type": "integer"
                    },
                    "min_n": {
                      "default": 10,
                      "description": "Minimum sample size (default 10)",
                      "type": "integer"
                    },
                    "top_k": {
                      "default": 15,
                      "description": "Top K combos to return (default 15)",
                      "type": "integer"
                    }
                  },
                  "type": "object"
                },
                "name": "find_signals"
              },
              {
                "description": "Compute \u03b1 stats for an arbitrary filter expression. Use this to test a specific hypothesis (e.g. 'tier_count_changed on enterprise-SaaS tickers' or 'severity 5 events that happened on Mondays'). Returns n, mean/median raw and \u03b1 returns at +1/+3/+7d, % positive, and the worst-loss trade.",
                "inputSchema": {
                  "properties": {
                    "co_occurrence_min": {
                      "description": "min same-day detector count (4 = 'real redesign')",
                      "type": "integer"
                    },
                    "detector": {
                      "description": "e.g. 'pricing_detector'",
                      "type": "string"
                    },
                    "event_type": {
                      "description": "e.g. 'TIER_COUNT_CHANGED' (case-insensitive)",
                      "type": "string"
                    },
                    "severity_min": {
                      "description": "minimum severity (1-5)",
                      "type": "integer"
                    },
                    "since": {
                      "description": "YYYY-MM-DD lower bound",
                      "type": "string"
                    },
                    "ticker": {
                      "description": "single ticker to filter to",
                      "type": "string"
                    },
                    "until": {
                      "description": "YYYY-MM-DD upper bound",
                      "type": "string"
                    }
                  },
                  "type": "object"
                },
                "name": "test_filter"
              },
              {
                "description": "Live signal feed: events fired in the last N days (default 7). Returns each event with the predicted \u03b1 range based on its event type's historical performance. Use this to surface 'what should I be looking at right now?'",
                "inputSchema": {
                  "properties": {
                    "days": {
                      "default": 7,
                      "description": "Lookback in calendar days (max 30)",
                      "type": "integer"
                    },
                    "min_co_occurrence": {
                      "description": "Only show events with this many same-day detectors (4 = high-conviction)",
                      "type": "integer"
                    }
                  },
                  "type": "object"
                },
                "name": "recent_events"
              },
              {
                "description": "Deep dive on a single event: full diff (added/removed values), surrounding price action (-3D to +14D), predicted vs actual \u03b1, links to wayback comparison. Use this to investigate a specific event flagged by find_signals or recent_events.",
                "inputSchema": {
                  "properties": {
                    "event_id": {
                      "description": "change_event id",
                      "type": "integer"
                    }
                  },
                  "required": [
                    "event_id"
                  ],
                  "type": "object"
                },
                "name": "event_dossier"
              },
              {
                "description": "Scan a URL as it appeared on a historical date via the Wayback Machine. Uses intel.boolsai.ai against the wayback-wrapped URL. Returns the same JSON shape as Boolsai Scan but for a historical snapshot. Use when investigating WHEN a vendor was added/removed.",
                "inputSchema": {
                  "properties": {
                    "date": {
                      "description": "YYYY-MM-DD \u2014 closest wayback snapshot on or before this date will be used",
                      "type": "string"
                    },
                    "url": {
                      "description": "Original URL (e.g. 'https://gymshark.com/')",
                      "type": "string"
                    }
                  },
                  "required": [
                    "url",
                    "date"
                  ],
                  "type": "object"
                },
                "name": "scan_at_date"
              },
              {
                "description": "All events fired on a single ticker, plus price action timeline. Use this to investigate one company's pattern (e.g. 'show me everything we caught on NFLX').",
                "inputSchema": {
                  "properties": {
                    "limit": {
                      "default": 50,
                      "type": "integer"
                    },
                    "ticker": {
                      "description": "e.g. 'NFLX'",
                      "type": "string"
                    }
                  },
                  "required": [
                    "ticker"
                  ],
                  "type": "object"
                },
                "name": "ticker_history"
              },
              {
                "description": "Run an SPY-benchmarked backtest on the WAYBACK historical event dataset (2+ years, 13K events) instead of the recent live event dataset (2 months, 1.7K events). Much bigger samples for statistical confidence. Group by change_type / key_path / domain.",
                "inputSchema": {
                  "properties": {
                    "exclude_noise": {
                      "default": true,
                      "description": "Filter out is_meta_noise=1 events",
                      "type": "boolean"
                    },
                    "group_by": {
                      "default": "key_path",
                      "description": "Dimension to slice on",
                      "enum": [
                        "change_type",
                        "key_path",
                        "key_name",
                        "parent_path",
                        "domain"
                      ],
                      "type": "string"
                    },
                    "horizon_days": {
                      "default": 7,
                      "description": "Forward-return window",
                      "type": "integer"
                    },
                    "min_n": {
                      "default": 20,
                      "description": "Minimum sample size",
                      "type": "integer"
                    },
                    "since": {
                      "description": "YYYY-MM-DD lower bound on event date (default: when prices start)",
                      "type": "string"
                    },
                    "top_k": {
                      "default": 15,
                      "type": "integer"
                    }
                  },
                  "type": "object"
                },
                "name": "wayback_backtest"
              },
              {
                "description": "Week-by-week wayback diff timeline for one domain. Returns every detected stack change (additions / removals) with week date. Use this to see when a vendor was added/removed historically, e.g. 'when did adobe.com add Segment?'",
                "inputSchema": {
                  "properties": {
                    "change_type": {
                      "default": "any",
                      "enum": [
                        "added",
                        "removed",
                        "changed",
                        "any"
                      ],
                      "type": "string"
                    },
                    "contains": {
                      "description": "Filter to events whose key_path or key_name contains this string (e.g. 'segment')",
                      "type": "string"
                    },
                    "domain": {
                      "description": "e.g. 'adobe.com'",
                      "type": "string"
                    },
                    "limit": {
                      "default": 100,
                      "type": "integer"
                    }
                  },
                  "required": [
                    "domain"
                  ],
                  "type": "object"
                },
                "name": "domain_timeline"
              },
              {
                "description": "ONE-SHOT cross-signal sweep. Computes \u03b1-vs-SPY stats simultaneously across event_type, detector, diff_field, severity, AND co_occurrence dimensions \u2014 returns the full landscape in a single response. Use this FIRST when you want to see where signal lives without having to call find_signals N times. Stateless, pure D1, no rate-limit risk, ~1s response. Cached per arg set for sub-100ms repeated queries.",
                "inputSchema": {
                  "properties": {
                    "horizon_days": {
                      "default": 7,
                      "description": "Forward-return window (default 7)",
                      "type": "integer"
                    },
                    "min_n": {
                      "default": 20,
                      "description": "Sample-size floor per group",
                      "type": "integer"
                    },
                    "since": {
                      "description": "Optional YYYY-MM-DD lower bound on event date",
                      "type": "string"
                    },
                    "source": {
                      "default": "both",
                      "description": "Which event dataset to scan. 'live' = 1.7K recent. 'wayback' = 13K over 2 years. 'both' = run both and return side-by-side.",
                      "enum": [
                        "live",
                        "wayback",
                        "both"
                      ],
                      "type": "string"
                    },
                    "top_k_per_dim": {
                      "default": 8,
                      "description": "Top K results per dimension (default 8)",
                      "type": "integer"
                    }
                  },
                  "type": "object"
                },
                "name": "signal_landscape"
              },
              {
                "description": "Compare two signal patterns side-by-side. e.g. 'how does PRICING_TIERS_ADDED compare to VENDORS_DETECTED_CHANGED on the live dataset?' Returns \u03b1, %pos, sample size, worst/best trades for each, plus delta. Pure D1, fast.",
                "inputSchema": {
                  "properties": {
                    "horizon_days": {
                      "default": 7,
                      "type": "integer"
                    },
                    "signal_a": {
                      "description": "First filter (same shape as test_filter args)",
                      "type": "object"
                    },
                    "signal_b": {
                      "description": "Second filter",
                      "type": "object"
                    }
                  },
                  "required": [
                    "signal_a",
                    "signal_b"
                  ],
                  "type": "object"
                },
                "name": "signal_diff"
              },
              {
                "description": "Bulk-farm a domain's historical wayback snapshots into our index. Use this when you need backtest history on a domain we haven't already farmed (i.e. wayback_backtest / domain_timeline return no data for it). Hits CDX \u2192 samples weekly \u2192 parallel-scans up to 50 snapshots via intel.boolsai.ai \u2192 inserts into wayback_intel_profiles. After farming completes you can call wayback_backtest or domain_timeline on the domain immediately. Cost: ~30-60s wall time, ~50 intel scans.",
                "inputSchema": {
                  "properties": {
                    "domain": {
                      "description": "Bare domain, e.g. 'sweetgreen.com'",
                      "type": "string"
                    },
                    "max_snapshots": {
                      "default": 50,
                      "description": "Hard cap on snapshots to fetch (default 50; max 200)",
                      "type": "integer"
                    },
                    "weeks": {
                      "default": 26,
                      "description": "How many weeks of history to farm (default 26 = ~6 months; max 100)",
                      "type": "integer"
                    }
                  },
                  "required": [
                    "domain"
                  ],
                  "type": "object"
                },
                "name": "farm_domain"
              }
            ]
          }
        },
        "bad_protocol_status_code": 200,
        "delete_error": null,
        "delete_status_code": null,
        "expired_session_error": null,
        "expired_session_status_code": null,
        "issues": [
          "missing_session_id",
          "missing_protocol_header",
          "bad_protocol_not_rejected"
        ],
        "last_event_id_visible": false,
        "protocol_header_present": false,
        "requested_protocol_version": "2025-03-26",
        "session_id_present": false,
        "transport": "streamable-http"
      },
      "latency_ms": 14.34,
      "status": "error"
    },
    "utility_coverage_probe": {
      "details": {
        "completions": {
          "advertised": false,
          "live_probe": "not_executed",
          "sample_target": null
        },
        "initialize_capability_keys": [
          "tools"
        ],
        "pagination": {
          "metadata_signal": false,
          "next_cursor_methods": [],
          "supported": false
        },
        "tasks": {
          "advertised": false,
          "http_status": 200,
          "probe_status": "missing"
        }
      },
      "latency_ms": 11.83,
      "status": "missing"
    }
  },
  "failures": {
    "oauth_authorization_server": {
      "reason": "no_authorization_server"
    },
    "oauth_protected_resource": {
      "error": "Client error '404 Not Found' for url 'https://signals.boolsai.ai/.well-known/oauth-protected-resource'\nFor more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/404",
      "url": "https://signals.boolsai.ai/.well-known/oauth-protected-resource"
    },
    "openid_configuration": {
      "reason": "no_authorization_server"
    },
    "server_card": {
      "error": "Client error '404 Not Found' for url 'https://signals.boolsai.ai/.well-known/mcp/server-card.json'\nFor more information check: https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/404",
      "url": "https://signals.boolsai.ai/.well-known/mcp/server-card.json"
    },
    "transport_compliance_probe": {
      "bad_protocol_error": null,
      "bad_protocol_headers": {
        "content-type": "application/json"
      },
      "bad_protocol_payload": {
        "id": 410,
        "jsonrpc": "2.0",
        "result": {
          "tools": [
            {
              "description": "Orient the agent: total events, tickers, date range, top event types, top detectors, price coverage, SPY benchmark status. Call this FIRST when starting research. Returns counts that let the agent reason about sample sizes before drilling in.",
              "inputSchema": {
                "properties": {},
                "type": "object"
              },
              "name": "universe_summary"
            },
            {
              "description": "Automated pattern discovery \u2014 scans event_type \u00d7 detector \u00d7 diff_field \u00d7 severity combinations and returns those with the strongest forward-return characteristics (\u03b1 vs SPY, % positive, n). Use this when you don't have a specific hypothesis yet. Returns sorted by \u03b1 at +7D descending. Filter by min_n to set a sample-size floor.",
              "inputSchema": {
                "properties": {
                  "group_by": {
                    "default": "event_type",
                    "description": "What dimension to slice on",
                    "enum": [
                      "event_type",
                      "detector",
                      "diff_field",
                      "severity",
                      "co_occurrence"
                    ],
                    "type": "string"
                  },
                  "horizon_days": {
                    "default": 7,
                    "description": "Forward-return window (default 7)",
                    "type": "integer"
                  },
                  "min_n": {
                    "default": 10,
                    "description": "Minimum sample size (default 10)",
                    "type": "integer"
                  },
                  "top_k": {
                    "default": 15,
                    "description": "Top K combos to return (default 15)",
                    "type": "integer"
                  }
                },
                "type": "object"
              },
              "name": "find_signals"
            },
            {
              "description": "Compute \u03b1 stats for an arbitrary filter expression. Use this to test a specific hypothesis (e.g. 'tier_count_changed on enterprise-SaaS tickers' or 'severity 5 events that happened on Mondays'). Returns n, mean/median raw and \u03b1 returns at +1/+3/+7d, % positive, and the worst-loss trade.",
              "inputSchema": {
                "properties": {
                  "co_occurrence_min": {
                    "description": "min same-day detector count (4 = 'real redesign')",
                    "type": "integer"
                  },
                  "detector": {
                    "description": "e.g. 'pricing_detector'",
                    "type": "string"
                  },
                  "event_type": {
                    "description": "e.g. 'TIER_COUNT_CHANGED' (case-insensitive)",
                    "type": "string"
                  },
                  "severity_min": {
                    "description": "minimum severity (1-5)",
                    "type": "integer"
                  },
                  "since": {
                    "description": "YYYY-MM-DD lower bound",
                    "type": "string"
                  },
                  "ticker": {
                    "description": "single ticker to filter to",
                    "type": "string"
                  },
                  "until": {
                    "description": "YYYY-MM-DD upper bound",
                    "type": "string"
                  }
                },
                "type": "object"
              },
              "name": "test_filter"
            },
            {
              "description": "Live signal feed: events fired in the last N days (default 7). Returns each event with the predicted \u03b1 range based on its event type's historical performance. Use this to surface 'what should I be looking at right now?'",
              "inputSchema": {
                "properties": {
                  "days": {
                    "default": 7,
                    "description": "Lookback in calendar days (max 30)",
                    "type": "integer"
                  },
                  "min_co_occurrence": {
                    "description": "Only show events with this many same-day detectors (4 = high-conviction)",
                    "type": "integer"
                  }
                },
                "type": "object"
              },
              "name": "recent_events"
            },
            {
              "description": "Deep dive on a single event: full diff (added/removed values), surrounding price action (-3D to +14D), predicted vs actual \u03b1, links to wayback comparison. Use this to investigate a specific event flagged by find_signals or recent_events.",
              "inputSchema": {
                "properties": {
                  "event_id": {
                    "description": "change_event id",
                    "type": "integer"
                  }
                },
                "required": [
                  "event_id"
                ],
                "type": "object"
              },
              "name": "event_dossier"
            },
            {
              "description": "Scan a URL as it appeared on a historical date via the Wayback Machine. Uses intel.boolsai.ai against the wayback-wrapped URL. Returns the same JSON shape as Boolsai Scan but for a historical snapshot. Use when investigating WHEN a vendor was added/removed.",
              "inputSchema": {
                "properties": {
                  "date": {
                    "description": "YYYY-MM-DD \u2014 closest wayback snapshot on or before this date will be used",
                    "type": "string"
                  },
                  "url": {
                    "description": "Original URL (e.g. 'https://gymshark.com/')",
                    "type": "string"
                  }
                },
                "required": [
                  "url",
                  "date"
                ],
                "type": "object"
              },
              "name": "scan_at_date"
            },
            {
              "description": "All events fired on a single ticker, plus price action timeline. Use this to investigate one company's pattern (e.g. 'show me everything we caught on NFLX').",
              "inputSchema": {
                "properties": {
                  "limit": {
                    "default": 50,
                    "type": "integer"
                  },
                  "ticker": {
                    "description": "e.g. 'NFLX'",
                    "type": "string"
                  }
                },
                "required": [
                  "ticker"
                ],
                "type": "object"
              },
              "name": "ticker_history"
            },
            {
              "description": "Run an SPY-benchmarked backtest on the WAYBACK historical event dataset (2+ years, 13K events) instead of the recent live event dataset (2 months, 1.7K events). Much bigger samples for statistical confidence. Group by change_type / key_path / domain.",
              "inputSchema": {
                "properties": {
                  "exclude_noise": {
                    "default": true,
                    "description": "Filter out is_meta_noise=1 events",
                    "type": "boolean"
                  },
                  "group_by": {
                    "default": "key_path",
                    "description": "Dimension to slice on",
                    "enum": [
                      "change_type",
                      "key_path",
                      "key_name",
                      "parent_path",
                      "domain"
                    ],
                    "type": "string"
                  },
                  "horizon_days": {
                    "default": 7,
                    "description": "Forward-return window",
                    "type": "integer"
                  },
                  "min_n": {
                    "default": 20,
                    "description": "Minimum sample size",
                    "type": "integer"
                  },
                  "since": {
                    "description": "YYYY-MM-DD lower bound on event date (default: when prices start)",
                    "type": "string"
                  },
                  "top_k": {
                    "default": 15,
                    "type": "integer"
                  }
                },
                "type": "object"
              },
              "name": "wayback_backtest"
            },
            {
              "description": "Week-by-week wayback diff timeline for one domain. Returns every detected stack change (additions / removals) with week date. Use this to see when a vendor was added/removed historically, e.g. 'when did adobe.com add Segment?'",
              "inputSchema": {
                "properties": {
                  "change_type": {
                    "default": "any",
                    "enum": [
                      "added",
                      "removed",
                      "changed",
                      "any"
                    ],
                    "type": "string"
                  },
                  "contains": {
                    "description": "Filter to events whose key_path or key_name contains this string (e.g. 'segment')",
                    "type": "string"
                  },
                  "domain": {
                    "description": "e.g. 'adobe.com'",
                    "type": "string"
                  },
                  "limit": {
                    "default": 100,
                    "type": "integer"
                  }
                },
                "required": [
                  "domain"
                ],
                "type": "object"
              },
              "name": "domain_timeline"
            },
            {
              "description": "ONE-SHOT cross-signal sweep. Computes \u03b1-vs-SPY stats simultaneously across event_type, detector, diff_field, severity, AND co_occurrence dimensions \u2014 returns the full landscape in a single response. Use this FIRST when you want to see where signal lives without having to call find_signals N times. Stateless, pure D1, no rate-limit risk, ~1s response. Cached per arg set for sub-100ms repeated queries.",
              "inputSchema": {
                "properties": {
                  "horizon_days": {
                    "default": 7,
                    "description": "Forward-return window (default 7)",
                    "type": "integer"
                  },
                  "min_n": {
                    "default": 20,
                    "description": "Sample-size floor per group",
                    "type": "integer"
                  },
                  "since": {
                    "description": "Optional YYYY-MM-DD lower bound on event date",
                    "type": "string"
                  },
                  "source": {
                    "default": "both",
                    "description": "Which event dataset to scan. 'live' = 1.7K recent. 'wayback' = 13K over 2 years. 'both' = run both and return side-by-side.",
                    "enum": [
                      "live",
                      "wayback",
                      "both"
                    ],
                    "type": "string"
                  },
                  "top_k_per_dim": {
                    "default": 8,
                    "description": "Top K results per dimension (default 8)",
                    "type": "integer"
                  }
                },
                "type": "object"
              },
              "name": "signal_landscape"
            },
            {
              "description": "Compare two signal patterns side-by-side. e.g. 'how does PRICING_TIERS_ADDED compare to VENDORS_DETECTED_CHANGED on the live dataset?' Returns \u03b1, %pos, sample size, worst/best trades for each, plus delta. Pure D1, fast.",
              "inputSchema": {
                "properties": {
                  "horizon_days": {
                    "default": 7,
                    "type": "integer"
                  },
                  "signal_a": {
                    "description": "First filter (same shape as test_filter args)",
                    "type": "object"
                  },
                  "signal_b": {
                    "description": "Second filter",
                    "type": "object"
                  }
                },
                "required": [
                  "signal_a",
                  "signal_b"
                ],
                "type": "object"
              },
              "name": "signal_diff"
            },
            {
              "description": "Bulk-farm a domain's historical wayback snapshots into our index. Use this when you need backtest history on a domain we haven't already farmed (i.e. wayback_backtest / domain_timeline return no data for it). Hits CDX \u2192 samples weekly \u2192 parallel-scans up to 50 snapshots via intel.boolsai.ai \u2192 inserts into wayback_intel_profiles. After farming completes you can call wayback_backtest or domain_timeline on the domain immediately. Cost: ~30-60s wall time, ~50 intel scans.",
              "inputSchema": {
                "properties": {
                  "domain": {
                    "description": "Bare domain, e.g. 'sweetgreen.com'",
                    "type": "string"
                  },
                  "max_snapshots": {
                    "default": 50,
                    "description": "Hard cap on snapshots to fetch (default 50; max 200)",
                    "type": "integer"
                  },
                  "weeks": {
                    "default": 26,
                    "description": "How many weeks of history to farm (default 26 = ~6 months; max 100)",
                    "type": "integer"
                  }
                },
                "required": [
                  "domain"
                ],
                "type": "object"
              },
              "name": "farm_domain"
            }
          ]
        }
      },
      "bad_protocol_status_code": 200,
      "delete_error": null,
      "delete_status_code": null,
      "expired_session_error": null,
      "expired_session_status_code": null,
      "issues": [
        "missing_session_id",
        "missing_protocol_header",
        "bad_protocol_not_rejected"
      ],
      "last_event_id_visible": false,
      "protocol_header_present": false,
      "requested_protocol_version": "2025-03-26",
      "session_id_present": false,
      "transport": "streamable-http"
    }
  },
  "remote_url": "https://signals.boolsai.ai/mcp",
  "server_card_payload": null,
  "server_identifier": "ai.boolsai/signals"
}