Tool Reference

Every tool the Mythic MCP server exposes — 313 tools across 31 scope families — with the scope each one needs and the API route it proxies.

About this reference

All 313 tools the Settings Server registers, grouped by the scope family that grants them. Each tool is a thin proxy over the HTTP API: the Proxies column names the route it calls, so anything you can do here you can also do over HTTP, and both answer from the same implementation.

Tools are gated by the scopes on your mcp_ key. Only tools within the key's scopes are listed; calling any other tool by name answers insufficient_scope with the scope to request, so an agent can tell you exactly what to add. A :write scope implies the matching :read. The same scope authorizes the Proxies route when the key is used directly over HTTP.

Client-scoped tools operate on the selected client — see the select-a-client flow.

Session

No scope needed. list_clients and select_client register for every agency-wide key (list_clients also fronts the roster route over HTTP, where it takes agency:read); whoami and search_docs register for every key.

Tool	Scope	What it does	Proxies
`list_clients` List Clients	—	List the clients (locations) this agency key can operate on.	`GET /builder/locations` `GET /client/v1/agency/locations`
`select_client` Select Client	—	Set the active client for this session by id (from list_clients).	—
`whoami` Who Am I	—	Describe this connection: key type (agency-wide or bound to one client), the pinned client if any, the scopes held, and per family whether this key can read, write, or nothing.	—
`search_docs` Search Documentation	—	Search the public Mythic Analytics documentation (docs.mythicdata.io): API routes, tool behaviour, scopes, error codes, SDK setup.	—

People

Person profiles, identities, merges and merge warnings.

Tool	Scope	What it does	Proxies
`list_people` List People	`people:read`	Resolved person profiles for this client, most recently seen first — display name, email, company, session counts, first/last channel and UTMs per row, so rows are labelable without a per-person fetch.	`GET /client/v1/data/people`
`get_person` Get Person	`people:read`	One full person profile: properties, all linked identities, first/last seen, session and device counts, first-touch geo/browser/OS.	`GET /client/v1/data/people/:personId`
`get_person_summary` Get Person Summary	`people:read`	Compact profile card for one person: properties, identities, first/last seen, plus a `metrics` block (session/device counts, first-touch landing page, UTMs, referrer, geo).	`GET /client/v1/data/people/:personId/summary`
`get_person_activity` Get Person Activity	`people:read`	One person's raw event stream, newest first.	`GET /client/v1/data/people/:personId/activity`
`get_person_attribution` Get Person Attribution	`people:read`	One person's marketing attribution: first-touch and last-touch channel/UTMs/referrer with first/last seen timestamps.	`GET /client/v1/data/people/:personId/attribution`
`list_identities` List Identities	`people:read`	Bulk identity mappings (identity_value -> canonical person_id) for agencies maintaining their own identity graph.	`GET /client/v1/data/identities`
`list_merges` List Person Merges	`people:read`	Authoritative person-merge history for this client (source_person_id merged into target_person_id, with reason and timestamp), oldest first.	`GET /client/v1/data/merges`
`get_person_merges` Get Person Merges	`people:read`	Merge history touching ONE person — every merge where they were the source or the target, oldest first.	`GET /client/v1/data/people/:personId/merges`
`get_person_logs` Get Person Logs	`people:read` + `logs:read`	One person's server log records across ALL of their identities (every distinct_id, user_id and email the identity graph links to them, up to 50; meta.identities_omitted > 0 says some were left out).	`GET /client/v1/data/people/:personId/logs`
`list_merge_warnings` List Merge Warnings	`people:read`	Merges the identity processor REFUSED or flagged as suspicious — each row names the event, the identity value, and the reason.	`GET /client/v1/data/merge-warnings`

Events & Explore

Event volumes, trends, breakdowns, property discovery, self-serve trend/funnel queries, tracking coverage and Core Web Vitals.

Tool	Scope	What it does	Proxies
`list_events` List Events	`events:read`	Every event name this client actually sends, with volumes, over a date range.	`GET /client/v1/data/events/summary`
`get_event_volume_daily` Get Daily Event Volume	`events:read`	Event counts per day (optionally for one event name via `event_type`).	`GET /client/v1/data/events/daily`
`get_event_trend` Get Event Trend	`events:read`	Overall event-volume trend over a window — defaults to the last 14 days (`days`), or pass explicit start_date/end_date.	`GET /client/v1/data/events/trend`
`get_event_breakdown` Get Event Breakdown	`events:read`	Traffic volume broken down by ONE dimension — top pages, referrers, devices, geo, or UTM values.	`GET /client/v1/data/events/breakdown`
`list_event_properties` List Event Properties	`events:read`	Property catalog for this client: every event-property or person-property key ever observed, with an inferred type, a masked sample value, how often it appears and first/last seen.	`GET /client/v1/data/properties`
`query_events` Query Events (trend / funnel / paths / attribution)	`events:read`	Self-serve analytics over the event stream in ONE call, sub-second: a TREND (1–5 series, each an event × math such as total/dau/unique_sessions/sum/p95, bucketed by hour/day/week/month or as a single number,…	`POST /client/v1/data/query`
`list_property_values` List Property Values	`events:read`	Top values of ONE event property over the last N days (default 30), with counts — the picker behind query_events filters and breakdowns.	`GET /client/v1/data/query/values`
`get_commerce_summary` Get Commerce Summary	`events:read`	Daily ecommerce totals — orders, revenue, average order value, units, buyers — for clients emitting the Mythic commerce spec (Shopify does automatically).	`GET /client/v1/data/commerce/summary`
`get_commerce_items` Get Commerce Items	`events:read`	Top products by revenue, units or orders — grouped by product id, SKU, variant, name, category or brand.	`GET /client/v1/data/commerce/items`
`explain_tracking_coverage` Explain Tracking Coverage	`events:read`	Can a "why is this page losing people" question be answered from the data we have for it? Run this BEFORE any behavioural analysis of a page and take the verdict at face value — it reports session volume, wh…	`GET /client/v1/data/explain/coverage`
`get_core_web_vitals` Get Core Web Vitals	`events:read`	Core Web Vitals measured on this client's site (lcp, fcp, cls, ttfb, fid, inp).	`GET /client/v1/performance/cwv`

Sessions

Session analytics — durations, engagement, breakdowns.

Tool	Scope	What it does	Proxies
`list_sessions` List Sessions	`sessions:read`	One row per website session, most recent first — landing/exit page, duration, pageviews, device/browser/geo, UTMs, engagement and bounce flags, identity link (distinct_id plus person_id, the canonical person…	`GET /client/v1/data/sessions`
`get_sessions_breakdown` Get Sessions Breakdown	`sessions:read`	Session aggregates grouped by ONE dimension — sessions, engagement, bounce rate etc.	`GET /client/v1/data/sessions/breakdown`
`get_session_health` Get Session Health	`sessions:read`	Checks whether the site's own code keeps starting new sessions.	`GET /client/v1/data/sessions/health`
`get_bot_traffic` Get Ad Platform Bot Traffic	`sessions:read`	Daily count of ad platform review bots Mythic dropped before storing anything: requests from Meta's, TikTok's, LinkedIn's or X's own networks, and declared ad crawlers such as AdsBot-Google.	`GET /client/v1/data/sessions/bot-traffic`

Session Replays

Compact, LLM-ready timelines of recorded visitor sessions (no PII).

Tool	Scope	What it does	Proxies
`list_replays` List Replays	`replays:read`	Index of sessions that have a screen recording, with interaction counters merged in where available (duration, clicks, keypresses, console_errors, network_count, last_url).	`GET /client/v1/data/replays`
`get_session_timeline` Get Session Timeline	`replays:read`	Compact, chronological timeline of one visitor session for the active client — page journey, dwell time, form submissions, field inputs, identify, custom events and errors.	`GET /client/v1/data/replays/:sessionId/timeline`
`get_session_logs` Get Session Logs	`replays:read` + `logs:read`	One visitor session's server log records: records the producer sent with this session's id as mythic.session_id (the browser's Mythic session id, forwarded with each request).	`GET /client/v1/data/replays/:sessionId/logs`

Error Tracking

Grouped JavaScript exceptions, occurrences and digests.

Tool	Scope	What it does	Proxies
`list_exception_issues` List Exception Issues	`exceptions:read`	Grouped JavaScript error issues — one row per fingerprinted issue (stable issue_id from error type + normalized message) with occurrence/session/user counts and a volume sparkline.	`GET /client/v1/data/exceptions/issues`
`get_exception_issue` Get Exception Issue	`exceptions:read`	Detail header for ONE grouped error issue by its stable issue_id (from list_exception_issues): counts, first/last seen and volume series over the window.	`GET /client/v1/data/exceptions/issues/:id`
`list_exception_occurrences` List Exception Occurrences	`exceptions:read`	Raw occurrences of ONE error issue — individual events with parsed stack frames, page URL and session_id, newest first.	`GET /client/v1/data/exceptions/issues/:id/occurrences`
`get_exception_digest` Get Exception Digest	`exceptions:read`	Weekly error digest in one call: summary with week-over-week deltas, a daily bar series, the top issues, and issues NEW this week (new = not seen in the preceding `new_lookback_days`, default 90).	`GET /client/v1/data/exceptions/digest`

Logs

OpenTelemetry log records sent over OTLP: search, record detail, volume, services, attributes, patterns and a bounded digest.

Tool	Scope	What it does	Proxies
`search_logs` Search Logs	`logs:read`	A client's log records, newest first, filtered by level, service, body text, attributes, distinct_id, session_id or trace_id.	`GET /client/v1/logs`
`get_log_record` Get Log Record	`logs:read`	One log record by uuid (from search_logs, a pattern sample, or the digest) with its FULL body, attributes and resource attributes.	`GET /client/v1/logs/records/:uuid`
`get_log_sparkline` Get Log Volume	`logs:read`	Log record counts per time bucket, split by level and zero-filled, for the filtered window.	`GET /client/v1/logs/sparkline`
`list_log_services` List Log Services	`logs:read`	Services (OTel service.name) that sent logs in the window, with record, error and warning counts and last-seen time.	`GET /client/v1/logs/services`
`list_log_attributes` List Log Attributes	`logs:read`	Attribute keys present on log records in the window, with how many records carry each.	`GET /client/v1/logs/attributes`
`list_log_attribute_values` List Log Attribute Values	`logs:read`	Top values of ONE attribute key in the window with counts.	`GET /client/v1/logs/attributes/values`
`list_log_patterns` List Log Patterns	`logs:read`	Log lines grouped into patterns: bodies that differ only in volatile tokens (dates, UUIDs, hex ids, numbers of 4+ digits; short numbers like status codes are kept) share a pattern_id.	`GET /client/v1/logs/patterns`
`get_log_pattern` Get Log Pattern	`logs:read`	One pattern by pattern_id: template, counts, first/last seen, sample, and its volume series over the window.	`GET /client/v1/logs/patterns/:id`
`list_log_pattern_occurrences` List Log Pattern Occurrences	`logs:read`	The records of ONE pattern, paged exactly like search_logs (next_cursor).	`GET /client/v1/logs/patterns/:id/occurrences`
`get_log_digest` Get Log Digest	`logs:read`	START HERE for logs.	`GET /client/v1/logs/digest`

Heatmaps

Click cells and scroll depth per page.

Tool	Scope	What it does	Proxies
`get_heatmap` Get Heatmap	`heatmaps:read`	Aggregated click grid for one page.	`GET /client/v1/data/heatmap`
`list_heatmap_pages` List Heatmap Pages	`heatmaps:read`	Pages that actually have heatmap data for this client, busiest first — START HERE before get_heatmap or get_heatmap_scroll.	`GET /client/v1/data/heatmap/pages`
`get_heatmap_facets` Get Heatmap Facets	`heatmaps:read`	Filter values that exist for one page (device_type, utm_source, utm_medium, and `viewport` layout buckets mobile/tablet/desktop with their px range), each with session counts, scoped to sessions that actuall…	`GET /client/v1/data/heatmap/facets`
`get_heatmap_scroll` Get Heatmap Scroll Depth	`heatmaps:read`	Cumulative scroll-depth curve for one page: bands of `band` px with how many sessions reached at least that depth.	`GET /client/v1/data/heatmap/scroll`

BigQuery Export

Export status and schema for the agency or one location.

Tool	Scope	What it does	Proxies
`get_bigquery_export_schema` Get BigQuery Export Schema	`bigquery_export:read`	The static semantic catalog of the BigQuery export: every table with its columns, types and descriptions.	`GET /client/v1/data/export/schema`
`get_bigquery_export` Get BigQuery Export	`bigquery_export:read`	This client's BigQuery export handle: the fully-qualified dataset (`mythic-data.<name>`), its table ids, provisioning status and timestamps.	`GET /client/v1/data/export`

Insights

Saved analytics queries and their results.

Tool	Scope	What it does	Proxies
`list_insights` List Insights	`insights:read`	List insights for the active client.	`GET /builder/insights`
`get_insight` Get Insight	`insights:read`	Get a single insight by ID, including its full config (config.query is the explore spec it runs).	`GET /builder/insights/:id`
`run_insight` Run Insight	`insights:read`	Execute a STORED insight and return its series/buckets (trend), steps (funnel) or rows/totals (attribution) — the same shape query_events returns — or, for an analysis panel, {engine:"analysis", kind, result…	`POST /builder/insights/:id/query`
`preview_insight` Preview Insight	`insights:read`	Execute an insight CONFIG without saving it — the same { insight_type, config } shape create_insight stores, run in-memory and never persisted.	`POST /builder/insights/preview`
`create_insight` Create Insight	`insights:write`	Create a new insight (a dashboard panel) for the active client.	`POST /builder/insights`
`update_insight` Update Insight	`insights:write`	Update an existing insight.	`PATCH /builder/insights/:id`
`delete_insight` Delete Insight	`insights:write`	Permanently delete an insight by ID.	`DELETE /builder/insights/:id`

Dashboards

Collections of insights arranged into dashboards.

Tool	Scope	What it does	Proxies
`list_dashboards` List Dashboards	`dashboards:read`	List all dashboards, optionally filtered by location_id.	`GET /builder/dashboards`
`get_dashboard` Get Dashboard	`dashboards:read`	Get a dashboard by ID, including all its insight items with layout and insight details.	`GET /builder/dashboards/:id`
`create_dashboard` Create Dashboard	`dashboards:write`	Create a new dashboard for the active client.	`POST /builder/dashboards`
`update_dashboard` Update Dashboard	`dashboards:write`	Update dashboard metadata.	`PATCH /builder/dashboards/:id`
`delete_dashboard` Delete Dashboard	`dashboards:write`	Permanently delete a dashboard and all its insight associations.	`DELETE /builder/dashboards/:id`
`add_insight_to_dashboard` Add Insight to Dashboard	`dashboards:write`	Add an insight to a dashboard with responsive grid layout positions.	`POST /builder/dashboards/:id/insights`
`remove_insight_from_dashboard` Remove Insight from Dashboard	`dashboards:write`	Remove an insight from a dashboard.	`DELETE /builder/dashboard-insights/:id`

Canvases

Freeform node/edge canvases with attached events.

Tool	Scope	What it does	Proxies
`list_canvases` List Canvases	`canvases:read`	List all canvases (visual tag management configurations), optionally filtered by tenant_key.	`GET /builder/canvases`
`get_canvas` Get Canvas	`canvases:read`	Get a canvas with all its nodes, edges, node URLs, and node events.	`GET /builder/canvases/:id`
`create_canvas` Create Canvas	`canvases:write`	Create a new canvas for visual tag management for the active client.	`POST /builder/canvases`
`update_canvas` Update Canvas	`canvases:write`	Update canvas metadata.	`PATCH /builder/canvases/:id`
`delete_canvas` Delete Canvas	`canvases:write`	Permanently delete a canvas and all its nodes, edges, and events (cascading).	`DELETE /builder/canvases/:id`
`add_canvas_node` Add Canvas Node	`canvases:write`	Add a node (page, action, decision, or offsite) to a canvas.	—
`update_canvas_node` Update Canvas Node	`canvases:write`	Update a canvas node's properties.	—
`delete_canvas_node` Delete Canvas Node	`canvases:write`	Delete a canvas node and its associated URLs, events, and edges (cascading).	—
`add_canvas_edge` Add Canvas Edge	`canvases:write`	Connect two canvas nodes with an edge (defines flow between pages/actions).	—
`delete_canvas_edge` Delete Canvas Edge	`canvases:write`	Remove an edge between two canvas nodes.	—
`add_node_event` Add Node Event	`canvases:write`	Add a tracking event to a canvas node (e.g., fire a Meta Pixel event on page load).	`POST /builder/canvases/:id/nodes/:nodeId/events`
`delete_node_event` Delete Node Event	`canvases:write`	Remove a tracking event from a canvas node.	`DELETE /builder/canvases/:id/events/:eventId`

Segments

Stored person segments — definitions, live counts, membership.

Tool	Scope	What it does	Proxies
`list_segments` List Segments	`segments:read`	Every stored segment for this client — reusable audience definitions over person profiles.	`GET /builder/segments`
`get_segment` Get Segment	`segments:read`	One segment by id, including its full filter definition.	`GET /builder/segments/:id`
`get_segment_count` Get Segment Count	`segments:read`	Person count for a segment.	`GET /builder/segments/:id/count`
`list_segment_members` List Segment Members	`segments:read`	The people currently matching a segment, resolved live, with pagination.	`GET /builder/segments/:id/members`
`preview_segment` Preview Segment	`segments:read`	Person count for a draft definition before saving it.	`POST /builder/segments/preview`
`create_segment` Create Segment	`segments:write`	Create a segment from a filter definition.	`POST /builder/segments`
`update_segment` Update Segment	`segments:write`	Update a segment.	`PATCH /builder/segments/:id`
`delete_segment` Delete Segment	`segments:write`	Permanently delete a segment.	`DELETE /builder/segments/:id`

Constraint Finder

End-to-end paid funnel analysis and its stored benchmarks.

Tool	Scope	What it does	Proxies
`find_constraint` Find Funnel Constraint	`constraints:read`	End-to-end paid-funnel analysis (Meta spend → pixel sessions → conversions) per campaign/adset/ad: impressions→clicks (ad), clicks→unique click ids that became a visit (tracking; split into delivery/load whe…	`GET /client/v1/constraints`
`list_constraint_events` List Constraint-Eligible Events	`constraints:read`	Which events can serve as lead_event / conversion_event for the constraint finder, and — the part no other endpoint reports — what share of each actually carries a session.	`GET /client/v1/constraints/events`
`get_constraint_audit` Constraint Tracking Audit	`constraints:read`	Tracking hygiene per campaign: WHICH campaigns are measurable at all, and WHERE the ad clicks that reached no campaign went.	`GET /client/v1/constraints/audit`
`get_constraint_options` Constraint Finder Options	`constraints:read`	The constraint finder vocabulary — valid group_by/attribution values, benchmark steps and their precedence chain, defaults, and usage hints.	`GET /client/v1/constraints/options`
`get_constraint_benchmarks` Get Stored Constraint Benchmarks	`constraints:read`	The standing benchmark targets on file for this location: the account-wide default plus per-campaign targets, applied automatically to every find_constraint call.	`GET /client/v1/constraints/benchmarks`
`set_constraint_benchmarks` Set Constraint Benchmarks	`constraints:write`	Store standing benchmark targets for ONE scope — a campaign (by id or name) or, with campaign omitted, the account-wide default.	`PUT /client/v1/constraints/benchmarks`
`delete_constraint_benchmarks` Delete Constraint Benchmarks	`constraints:write`	Clear stored benchmark targets for ONE scope, or one step of it.	`DELETE /client/v1/constraints/benchmarks`

Correlation

Pairwise and rolling metric correlation with lead/lag scan.

Tool	Scope	What it does	Proxies
`get_correlation` Correlate Metrics	`correlation:read`	Trend of 2-5 metrics over day/week/month buckets with every pairwise Pearson AND Spearman coefficient, a rolling correlation folded into the series rows (`corr:<x>~<y>`, plottable on the same axis), and a le…	`GET /client/v1/correlation`
`list_correlation_metrics` List Correlatable Metrics	`correlation:read`	Real discovery of what can be correlated for THIS client, not a static list: event families list the event names actually captured in the window, Meta metrics are marked unavailable when the client has no ad…	`GET /client/v1/correlation/metrics`
`get_correlation_options` Correlation Options	`correlation:read`	The correlation vocabulary at no query cost: grains, the fixed metric registry, the metric-family grammar (family:<event_name>), and every query param with its bounds and defaults.	`GET /client/v1/correlation/options`

Incrementality

Geo-holdout experiments, iROAS and coefficients.

Tool	Scope	What it does	Proxies
`list_incrementality_markets` List Incrementality Markets	`incrementality:read`	Daily revenue by geographic region over a lookback, with candidate CONTROL markets ranked by Pearson correlation against a proposed test set.	`GET /client/v1/incrementality/markets`
`get_incrementality_eligibility` Incrementality Eligibility	`incrementality:read`	Pre-flight before designing any geo experiment: geo coverage of revenue, history depth, matchable market count, and the daily budget the client's noise floor implies for a conclusive 30-day test.	`GET /client/v1/incrementality/eligibility`
`list_incrementality_experiments` List Incrementality Experiments	`incrementality:read`	Every geo-holdout experiment for this client with definition, design (power analysis), result, and derived lifecycle progress (pending/running/cooldown/complete).	`GET /client/v1/incrementality/experiments`
`get_incrementality_experiment` Get Incrementality Experiment	`incrementality:read`	One experiment by id: definition, design, stored result (normalized on read), and lifecycle progress.	`GET /client/v1/incrementality/experiments/:id`
`get_incrementality_launch_plan` Incrementality Launch Plan	`incrementality:read`	The exact Meta ad-side recipe for conducting the holdout: control regions resolved to Meta targeting keys, apply/revert dates, paste-ready API calls.	`GET /client/v1/incrementality/experiments/:id/launch-plan`
`validate_incrementality_experiment` Validate Incrementality Experiment	`incrementality:read`	Two modes, chosen by phase.	`GET /client/v1/incrementality/experiments/:id/validate`
`get_incrementality_coefficients` Get Incrementality Coefficients	`incrementality:read`	The progressive truth: computed experiments pooled by coefficient series (channel + revenue definition + spend scope — only same-estimand experiments pool) into an inverse-variance weighted median.	`GET /client/v1/incrementality/coefficients`
`get_incrementality_coefficients_history` Incrementality Coefficient History	`incrementality:read`	The progressive truth over time: every computed experiment's coefficient in sequence plus the running weighted median at each point, per coefficient series.	`GET /client/v1/incrementality/coefficients/history`
`create_incrementality_experiment` Create Incrementality Experiment	`incrementality:write`	Create a geo-holdout experiment (status draft).	`POST /client/v1/incrementality/experiments`
`delete_incrementality_experiment` Delete Incrementality Experiment	`incrementality:write`	Delete an experiment.	`DELETE /client/v1/incrementality/experiments/:id`
`compute_incrementality` Compute Incrementality	`incrementality:write`	Run the iROAS computation for an experiment: weighted-synthetic-control diff-in-diff over pre vs test+cooldown, folded with the Meta spend row, stored as the experiment result (status becomes computed).	`POST /client/v1/incrementality/experiments/:id/compute`

Event Contracts

Declared per-event property schemas and their violation reports.

Tool	Scope	What it does	Proxies
`list_event_contracts` List Event Contracts	`contracts:read`	The event contracts stored for this client — each one an event name plus the properties it is declared to carry, their types, and which are required.	`GET /client/v1/contracts`
`check_event_contracts` Check Event Contracts	`contracts:read`	Diff every stored contract (or one, with `event`) against live traffic and report violations, worst first.	`GET /client/v1/contracts/violations`
`get_contract_samples` Get Contract Violation Samples	`contracts:read`	The actual offending events behind ONE violation from check_event_contracts — up to 50 raw examples with session_id and $current_url, so you can see where the bad data comes from.	`GET /client/v1/contracts/samples`
`suggest_event_contract` Suggest Event Contract	`contracts:read`	Infer a contract for one event from its live traffic — the fastest way to author one.	`GET /client/v1/contracts/suggest`
`get_contract_history` Get Contract History	`contracts:read`	Daily results of this client's event contracts, as the ~08:00 UTC sweep judged each previous UTC day: status (ok / violations / no_data), event counts, lowest required-property coverage, per-property coverag…	`GET /client/v1/contracts/history`
`list_contract_candidates` List Contract Candidates	`contracts:read`	Which events this client should put a contract on first, ranked.	`GET /client/v1/contracts/candidates`
`list_contract_presets` List Contract Presets	`contracts:read`	The built-in contract packs and every contract in them.	`GET /client/v1/contracts/presets`
`set_event_contract` Set Event Contract	`contracts:write`	Create or replace the contract for ONE event.	`PUT /client/v1/contracts`
`delete_event_contract` Delete Event Contract	`contracts:write`	Remove one event contract.	`DELETE /client/v1/contracts/:event`
`apply_contract_preset` Apply Contract Preset	`contracts:write`	Apply a built-in contract pack to a client in one call.	`POST /client/v1/contracts/presets/:preset`

Destinations

Activation destinations events are forwarded to, their credentials and deliveries.

Tool	Scope	What it does	Proxies
`list_destinations` List Destinations	`destinations:read`	List activation destinations for the active client (webhook, Meta CAPI, Google Ads).	`GET /client/v1/destinations`
`get_destination` Get Destination	`destinations:read`	Get a single activation destination by ID, including its full action_config and trigger_rules.	`GET /client/v1/destinations/:id`
`list_destination_credentials` List Destination Credentials	`destinations:read`	Which destination credentials are stored for this client — METADATA ONLY (id, destination_type, credential_name, created/updated timestamps).	`GET /client/v1/destinations/credentials`
`get_destination_oauth_client` Get Destination OAuth Client	`destinations:read`	Which Google OAuth client this agency's consent flows run as: has_client (agency registered its own), source ('agency' or the platform default), the client_id (not a secret — what you check against the Cloud…	`GET /client/v1/destinations/oauth/client`
`list_destination_deliveries` List Destination Deliveries	`destinations:read`	Delivery log for this client's destinations, newest first — one row per delivery attempt with status (success\|failed\|retrying\|skipped\|test), response status, error, and timing.	`GET /client/v1/destinations/deliveries` `GET /client/v1/destinations/:id/deliveries`
`list_destination_activity` List Destination Activity	`destinations:read`	Activity log for this client's destinations, newest first: ONE row per (source event x destination) with the outcome of its LATEST attempt (success\|failed\|skipped\|retrying; test rows only with status=test),…	`GET /client/v1/destinations/activity`
`get_destination_delivery` Get Destination Delivery	`destinations:read`	Full detail for ONE delivery (destination + source event uuid from list_destination_deliveries): the request actually sent (body + headers), the destination's response, attempt counts, field warnings/violati…	`GET /client/v1/destinations/:id/deliveries/:eventUuid`
`get_destination_delivery_status` Get Destination Delivery Status	`destinations:read`	What Google did with a delivery AFTER accepting it — google_ads / google_analytics destinations only.	`GET /client/v1/destinations/:id/deliveries/:eventUuid/status`
`get_destination_health` Get Destination Health	`destinations:read`	Per-destination delivery health rollup over a recent window: success/failure/skip counts, success rate, latency (avg + p95), queue lag, last error, and a credential_ok flag (false when the most recent failur…	`GET /client/v1/destinations/health`
`verify_destination` Verify Destination	`destinations:read`	Did deliveries actually COUNT downstream? convert_experiments destinations only.	`GET /client/v1/destinations/:id/verification`
`get_mida_experiments` Get Mida Experiments	`destinations:read`	Mida's OWN readout for this client's Mida (mida.so) project, plus how long each running test still needs.	`GET /client/v1/destinations/mida/experiments`
`get_destination_quality` Get Destination Quality	`destinations:read`	Meta's OWN verdict on the data this client sends — meta_capi destinations only.	`GET /client/v1/destinations/:id/quality`
`get_destination_quality_trend` Get Destination Quality Trend	`destinations:read`	Meta match quality for one destination DAY BY DAY — event match quality, server-vs-browser coverage, dedupe-key coverage on both halves, and the findings that stood on each day.	`GET /client/v1/destinations/:id/quality/trend`
`get_destination_quality_match_keys` Get Destination Match Key Coverage	`destinations:read`	Per-PARAMETER match coverage over time for one Meta destination: email, phone, fn, ln, fbp, fbc, external_id, ip_address, ct, st, zip, country — each as a daily percentage.	`GET /client/v1/destinations/:id/quality/match-keys`
`list_platform_integrations` List Platform Integrations	`destinations:read`	Show the managed Meta / Google Ads integrations for the active client: which halves are connected (browser pixel tag, server destination) and whether the delivery credential is stored.	`GET /client/v1/destinations/integrations`
`create_destination` Create Destination	`destinations:write`	Create an activation destination for the active client.	`POST /client/v1/destinations`
`update_destination` Update Destination	`destinations:write`	Update an existing destination.	`PATCH /client/v1/destinations/:id`
`delete_destination` Delete Destination	`destinations:write`	Permanently delete an activation destination by ID.	`DELETE /client/v1/destinations/:id`
`set_destination_credential` Set Destination Credential	`destinations:write`	Store (or rotate — same type+name overwrites) the credential one destination type delivers with.	`PUT /client/v1/destinations/credentials/:type`
`delete_destination_credential` Delete Destination Credential	`destinations:write`	Delete stored credential(s) for one destination type.	`DELETE /client/v1/destinations/credentials/:type`
`set_destination_oauth_client` Set Destination OAuth Client	`destinations:write`	Register the agency's OWN Google OAuth client for destination consent flows (their brand on the consent screen, their API quota) instead of the platform default.	`PUT /client/v1/destinations/oauth/client`
`delete_destination_oauth_client` Delete Destination OAuth Client	`destinations:write`	Remove the agency's registered Google OAuth client; future consent flows fall back to the platform default (Mythic's brand and quota).	`DELETE /client/v1/destinations/oauth/client`
`initiate_destination_oauth` Initiate Destination OAuth	`destinations:write`	Start the Google consent flow that stores a google_ads / google_analytics credential WITHOUT anyone handling a refresh token.	`POST /client/v1/destinations/oauth/initiate`
`test_destination` Test Destination	`destinations:write`	Fire a REAL test delivery through one destination — this sends an actual event to the live ad platform using the stored credential, and the attempt lands in the delivery log (status "test").	`POST /client/v1/destinations/:id/test`
`connect_meta_integration` Connect Meta Integration	`destinations:write`	One-click Meta tracking: creates the managed Meta Pixel browser tag AND the Meta CAPI server destination.	`POST /client/v1/destinations/integrations/meta`
`connect_google_ads_integration` Connect Google Ads Integration	`destinations:write`	One-click Google Ads tracking: creates the Google Ads (Data Manager) server destination, plus a managed gtag browser tag when any event has a conversion_label.	`POST /client/v1/destinations/integrations/google_ads`
`connect_google_analytics_integration` Connect GA4 Integration	`destinations:write`	One-click GA4 measurement with NO page tag: creates the GA4 (Data Manager) server destination mapped to the commerce spec.	`POST /client/v1/destinations/integrations/google_analytics`
`connect_openai_ads_integration` Connect OpenAI Ads Integration	`destinations:write`	One-click OpenAI Ads conversions: creates the OpenAI Ads server destination mapped from the commerce spec to OpenAIu2019s event taxonomy.	`POST /client/v1/destinations/integrations/openai_ads`
`delete_platform_integration` Delete Platform Integration	`destinations:write`	Disconnect a managed integration (Meta, Google Ads, GA4, OpenAI Ads): detaches the managed browser tag where there is one and removes the managed destination (and its stored credential when nothing else uses…	`DELETE /client/v1/destinations/integrations/:platform`

Ad UTM Auto-tagging

Meta ad UTM tagging for one client: config and campaign rules, the ad accounts its token can see, previews of which ads lack UTMs, applying the fix (creative copy + swap), the scheduled auto mode and run history.

Tool	Scope	What it does	Proxies
`get_meta_ad_tagging` Get Meta Ad UTM Tagging	`ad_tagging:read`	This client's Meta ad UTM tagging config (ad account, template, overwrite policy, campaign rules, auto mode), whether a meta_ads token is stored, the last apply/auto run, and whether live writes are switched…	`GET /client/v1/ad-tagging/meta`
`list_meta_ad_accounts` List Meta Ad Accounts	`ad_tagging:read`	The Meta ad accounts the stored meta_ads token can see (id, account_id, name, status, currency).	`GET /client/v1/ad-tagging/meta/ad-accounts`
`list_meta_ad_tagging_runs` List Meta Ad Tagging Runs	`ad_tagging:read`	Recent preview, apply and auto runs for this client, newest first, with counts (planned, applied, failed, uncertain, skipped, stale).	`GET /client/v1/ad-tagging/meta/runs`
`get_meta_ad_tagging_run` Get Meta Ad Tagging Run	`ad_tagging:read`	One run with its per-ad journal: url_tags before/after, the action, and the outcome (applied, reconciled, failed, uncertain, stale, skipped with reason).	`GET /client/v1/ad-tagging/meta/runs/:id`
`set_meta_ad_tagging` Set Meta Ad UTM Tagging	`ad_tagging:write`	Create or update this client's config, including how often automatic scans run (auto_interval_minutes).	`PUT /client/v1/ad-tagging/meta`
`delete_meta_ad_tagging` Delete Meta Ad UTM Tagging	`ad_tagging:write`	Remove the config and cancel its queued or running runs.	`DELETE /client/v1/ad-tagging/meta`
`preview_meta_ad_tagging` Preview Meta Ad UTM Tagging	`ad_tagging:write`	Read the matching campaigns, their ads and creatives from Meta and decide per ad: "set" (with url_tags before/after) or "skip" with a reason (already_correct, has_values_fill_missing, inline_utms_in_link, un…	`POST /client/v1/ad-tagging/meta/preview`
`apply_meta_ad_tagging` Apply Meta Ad UTM Tagging	`ad_tagging:write`	Queue the selected "set" ads from a preview of the SAVED config (made in the last 24h, tagging config unchanged since; interval-only edits are allowed).	`POST /client/v1/ad-tagging/meta/apply`
`set_meta_ad_tagging_auto` Set Meta Ad UTM Tagging Auto Mode	`ad_tagging:write`	Turns automatic scans on or off.	`POST /client/v1/ad-tagging/meta/auto`
`cancel_meta_ad_tagging_run` Cancel Meta Ad Tagging Run	`ad_tagging:write`	Stop a queued or running apply/auto run.	`POST /client/v1/ad-tagging/meta/runs/:id/cancel`

Tracking Settings

Per-location ingestion controls: allowed origins, autocapture, heatmaps, experiments, conversions, contract alerts — and the ingestion on/off switch.

Tool	Scope	What it does	Proxies
`get_tracking_settings` Get Tracking Settings	`settings:read`	Read this client's per-location tracking settings.	`GET /client/v1/settings/allowed-origins` `GET /client/v1/settings/autocapture` `GET /client/v1/settings/ip-geo-enrichment` `GET /client/v1/settings/heatmaps` `GET /client/v1/settings/experiments` `GET /client/v1/settings/conversions` `GET /client/v1/settings/session-replay` `GET /client/v1/settings/contract-alerts` `GET /client/v1/settings/timezone`
`get_client_status` Get Client Status	`settings:read`	Whether this client (location) is active — an inactive client's events are rejected at ingestion.	`GET /client/v1/status`
`update_tracking_setting` Update Tracking Setting	`settings:write`	Update ONE tracking setting for this client.	`PUT /client/v1/settings/allowed-origins` `PUT /client/v1/settings/autocapture` `PUT /client/v1/settings/ip-geo-enrichment` `PUT /client/v1/settings/heatmaps` `PUT /client/v1/settings/experiments` `PUT /client/v1/settings/conversions` `PUT /client/v1/settings/session-replay` `PUT /client/v1/settings/contract-alerts` `PUT /client/v1/settings/timezone`
`set_client_status` Set Client Status	`settings:write`	Activate or deactivate this client.	`POST /client/v1/status`

Client Provisioning

Create new clients under the agency. Agency-wide by nature: a client-scoped key must not be able to create siblings.

Agency-wide keys only — a key bound to one client cannot hold this scope.

Tool	Scope	What it does	Proxies
`create_location` Create Location	`provisioning:write`	Provision a new client (location) under your agency.	`POST /client/v1/locations`

Client Setup

One client’s own setup — details, install snippet, custom tracking hostname and install verification.

Tool	Scope	What it does	Proxies
`get_location` Get Location	`locations:read`	This client's provisioning details: id, name, publishable key, secret key, domain, contact fields, timezone, SDK global name, and whether tracking is currently enabled.	`GET /client/v1/locations/details`
`get_location_plan` Get Client Tier	`locations:read`	Which billing tier this client is on, and the tiers it can be moved to.	`GET /client/v1/locations/plan`
`get_location_loader` Get Install Snippet	`locations:read`	The tracking snippet to paste into this client's site, plus the publishable key, API host and SDK global name it uses.	`GET /client/v1/locations/loader`
`get_location_hostname` Get Custom Tracking Hostname	`locations:read`	This client's custom tracking hostname and its validation state (Cloudflare hostname status and SSL status), or null when none is attached.	`GET /client/v1/locations/hostname`
`get_ghl_connection` Get GoHighLevel Connection	`locations:read`	This client's GoHighLevel connection: status (active, not_connected, token_inactive, incomplete), the linked GHL sub-account id, and when it connected.	`GET /client/v1/locations/ghl`
`get_location_last_event` Check Tracking Is Live	`locations:read`	Event count and the most recent event timestamp for this client.	`GET /client/v1/locations/last-event`
`list_location_events` List Raw Events	`locations:read`	Individual events for this client, newest first — the debugging view for confirming an install or chasing a missing property.	`GET /client/v1/locations/events`
`update_location` Update Location	`locations:write`	Update a client's own details — name, domain, email, phone, website, timezone or global_name.	`PATCH /client/v1/locations/details`
`set_location_plan` Set Client Tier	`locations:write`	Move this client to a different billing tier.	`PUT /client/v1/locations/plan`
`set_location_hostname` Set Custom Tracking Hostname	`locations:write`	Attach a first-party tracking hostname (e.g.	`PUT /client/v1/locations/hostname`
`delete_location_hostname` Remove Custom Tracking Hostname	`locations:write`	Detach this client's custom tracking hostname and stop routing events for it.	`DELETE /client/v1/locations/hostname`
`create_ghl_install_link` Create GoHighLevel Install Link	`locations:write`	A link that installs the Mythic GoHighLevel app on one GHL sub-account and links it to this client.	`POST /client/v1/locations/ghl/install`

Links & Uptime

Link redirects, click analytics, link webhooks and uptime.

Tool	Scope	What it does	Proxies
`list_links` List Links	`links:read`	This client's link redirects, newest first, each enriched with its click_count for date_from..date_to (default: the last 30 days).	`GET /client/v1/links`
`get_link_analytics` Get Link Analytics	`links:read`	Aggregated click analytics per link (totals by slug), optionally narrowed to one slug, a date range, or specific UTM values.	`GET /client/v1/links/analytics`
`get_link_trend` Get Link Click Trend	`links:read`	Click counts over time for link redirects — the time-series companion to get_link_analytics.	`GET /client/v1/links/trend`
`get_click_id_recovery` Get Click ID Recovery	`links:read`	One row per ad platform (google, meta, microsoft): paid sessions, how many arrived without their click ID (gclid/gbraid/wbraid, fbclid, msclkid), and how many of those the Mythic short link restored from the…	`GET /client/v1/links/click-id-recovery`
`get_link_click_stream` Get Link Click Stream	`links:read`	Individual recent link clicks (real-time stream), newest first — who clicked what, with UTM detail.	`GET /client/v1/links/stream`
`list_link_utm_values` List Link UTM Values	`links:read`	Distinct UTM values seen on link clicks — use to discover what to pass as UTM filters to the analytics/trend/stream tools.	`GET /client/v1/links/utm-values`
`list_link_webhooks` List Link Uptime Webhooks	`links:read`	The uptime notification webhooks for this client — each fires on link_down and/or link_up when a monitored link destination goes down or recovers.	`GET /client/v1/links/webhooks`
`get_link_uptime` Get Link Uptime	`links:read`	Current up/down status of each monitored link destination (or one, with slug).	`GET /client/v1/links/uptime`
`get_link_uptime_history` Get Link Uptime History	`links:read`	Uptime check time-series for monitored link destinations — status codes and response times over a date range.	`GET /client/v1/links/uptime/history`
`create_link` Create Link	`links:write`	Create a link redirect.	`POST /client/v1/links`
`update_link` Update Link	`links:write`	Update fields on an existing link redirect (partial update — only the fields you pass change).	`PATCH /client/v1/links/:linkId`
`delete_link` Delete Link	`links:write`	Permanently delete a link redirect.	`DELETE /client/v1/links/:linkId`
`create_link_webhook` Create Link Uptime Webhook	`links:write`	Create a webhook that is called when a monitored link destination goes down (link_down) or recovers (link_up).	`POST /client/v1/links/webhooks`
`update_link_webhook` Update Link Uptime Webhook	`links:write`	Update an uptime webhook (partial — only the fields you pass change).	`PATCH /client/v1/links/webhooks/:webhookId`
`delete_link_webhook` Delete Link Uptime Webhook	`links:write`	Permanently delete an uptime webhook.	`DELETE /client/v1/links/webhooks/:webhookId`

Tool	Scope	What it does	Proxies
`list_tags` List Tags	`tags:read`	List tag assignments (tenant_tags) for the active client.	`GET /client/v1/config/tags`
`list_tag_templates` List Tag Templates	`tags:read`	Tag templates this client can assign with create_tag: Mythic's catalog (owner "mythic", read-only) plus the client's own (owner "location").	`GET /client/v1/config/tag-templates`
`get_tag_template` Get Tag Template	`tags:read`	One tag template by id, including its src or code.	`GET /client/v1/config/tag-templates/:id`
`get_tag_workspace` Get Tag Workspace	`tags:read`	Publishing mode (immediate or draft) and what a publish would change: every tag, snippet and transformer added, removed or modified since the latest version.	`GET /client/v1/config/workspace`
`list_tag_versions` List Tag Versions	`tags:read`	Version history of the tag config, newest first: number, source (auto, publish, restore, mode_change), name, notes, who created it and when.	`GET /client/v1/config/versions`
`get_tag_version` Get Tag Version	`tags:read`	One version: what changed from the previous version, the full snapshot (disabled rows included) and the layer the SDK runs from it.	`GET /client/v1/config/versions/:version`
`get_live_tag_config` Get Live Tag Config	`tags:read`	The tags, snippets and transformers the SDK is served right now, which version that is, and in_sync: whether it matches what should be live (latest version in draft mode, workspace in immediate mode).	`GET /client/v1/config/live`
`list_tag_firings` List Tag Firings	`tags:read`	How often each tag and snippet fired in a window: fires, sessions, distinct pages, distinct triggers, first and last fired.	`GET /client/v1/config/firings`
`get_tag_firing_breakdown` Get Tag Firing Breakdown	`tags:read`	Where one tag or snippet fired: by page (host + path), by trigger ("page_load" or the Mythic event that triggered it) or by day.	`GET /client/v1/config/firings/:kind/:id`
`create_tag` Create Tag Assignment	`tags:write`	Assign a tag to the active client (tenant_tags row).	`POST /client/v1/config/tags`
`update_tag` Update Tag Assignment	`tags:write`	Update a tag assignment for the active client.	`PATCH /client/v1/config/tags/:tagId`
`delete_tag` Delete Tag Assignment	`tags:write`	Remove a tag assignment from the active client.	`DELETE /client/v1/config/tags/:tagId`
`create_tag_template` Create Tag Template	`tags:write`	Create a tag template owned by this client.	`POST /client/v1/config/tag-templates`
`update_tag_template` Update Tag Template	`tags:write`	Update one of this client's own tag templates (Mythic's are read-only).	`PATCH /client/v1/config/tag-templates/:id`
`delete_tag_template` Delete Tag Template	`tags:write`	Delete one of this client's own tag templates.	`DELETE /client/v1/config/tag-templates/:id`
`set_tag_publish_mode` Set Tag Publish Mode	`tags:write`	immediate: every tag, snippet and transformer write goes live and records a version.	`PUT /client/v1/config/publish-mode`
`publish_tag_config` Publish Tag Config	`tags:write`	Record the workspace (tags, snippets, transformers) as a new version and make it live.	`POST /client/v1/config/publish`
`restore_tag_version` Restore Tag Version	`tags:write`	Roll back: rewrite the workspace to an earlier version and publish it as a new version.	`POST /client/v1/config/versions/:version/restore`

Snippets

Custom code snippets injected by the SDK.

Tool	Scope	What it does	Proxies
`list_snippets` List Snippets	`snippets:read`	List custom code snippets (tenant_snippets) for the active client.	`GET /client/v1/config/snippets`
`get_snippet` Get Snippet	`snippets:read`	Get a single code snippet by ID, including its full code.	`GET /client/v1/config/snippets/:id`
`create_snippet` Create Snippet	`snippets:write`	Create a custom code snippet for the active client.	`POST /client/v1/config/snippets`
`update_snippet` Update Snippet	`snippets:write`	Update a code snippet.	`PATCH /client/v1/config/snippets/:id`
`delete_snippet` Delete Snippet	`snippets:write`	Permanently delete a code snippet by ID.	`DELETE /client/v1/config/snippets/:id`

Transformers

Event transformers applied during ingestion.

Tool	Scope	What it does	Proxies
`list_transformers` List Transformers	`transformers:read`	List event enrichment transformers for the active client.	`GET /client/v1/config/transformers`
`get_transformer` Get Transformer	`transformers:read`	Get a single transformer by ID, including its full config.	`GET /client/v1/config/transformers/:id`
`create_transformer` Create Transformer	`transformers:write`	Create an event enrichment transformer for the active client.	`POST /client/v1/config/transformers`
`update_transformer` Update Transformer	`transformers:write`	Update a transformer.	`PATCH /client/v1/config/transformers/:id`
`delete_transformer` Delete Transformer	`transformers:write`	Permanently delete a transformer by ID.	`DELETE /client/v1/config/transformers/:id`

Airbyte Sync

Agency-grain sync orchestration: connections, schedules, pipelines, sync history.

Agency-wide keys only — a key bound to one client cannot hold this scope.

Tool	Scope	What it does	Proxies
`list_airbyte_platforms` List Airbyte Platforms	`airbyte:read`	The ad/CRM platforms Mythic can connect through Airbyte (google_ads, meta_ads, bing_ads, tiktok_ads, salesforce, …) with their default streams, sync options and whether they support the OAuth consent flow.	`GET /client/v1/airbyte/platforms`
`get_airbyte_config` Get Airbyte Config	`airbyte:read`	This agency's Airbyte workspace configuration and provisioning status.	`GET /client/v1/airbyte/config`
`list_airbyte_webhooks` List Airbyte Webhook Subscriptions	`airbyte:read`	The agency's sync-notification relay subscriptions.	`GET /client/v1/airbyte/webhooks`
`list_airbyte_connections` List Airbyte Connections	`airbyte:read`	Every Airbyte connection this agency owns, optionally filtered.	`GET /client/v1/airbyte/connections`
`get_airbyte_connection` Get Airbyte Connection	`airbyte:read`	One connection by its Mythic id — platform, client, enabled streams, schedule, sync_dow, last sync.	`GET /client/v1/airbyte/connections/:id`
`list_airbyte_jobs` List Airbyte Jobs	`airbyte:read`	A connection's recent sync jobs, fetched live from Airbyte.	`GET /client/v1/airbyte/connections/:id/jobs`
`list_airbyte_sync_events` List Airbyte Sync Events	`airbyte:read`	Sync history as received from Airbyte's notifications, newest first.	`GET /client/v1/airbyte/sync-events`
`get_airbyte_stats` Get Airbyte Stats	`airbyte:read`	Aggregate sync statistics across all of this agency’s connections (volumes, success rates).	`GET /client/v1/airbyte/stats`
`get_airbyte_health` Get Airbyte Health	`airbyte:read`	Per-connection health: latest sync plus the consecutive-failure streak.	`GET /client/v1/airbyte/health`
`get_airbyte_pipeline` Get Airbyte Pipeline	`airbyte:read`	One client's pipeline config: the connections in scope, enabled flag, run_hour_utc, and completion-webhook settings (the secret is never returned — only has_webhook_secret).	`GET /client/v1/airbyte/pipelines/:clientId`
`list_airbyte_pipeline_candidates` List Airbyte Pipeline Candidates	`airbyte:read`	This client's connections annotated with in_scope (already in the pipeline) and suggested (currently active — a reasonable default).	`GET /client/v1/airbyte/pipelines/:clientId/candidates`
`list_airbyte_pipeline_runs` List Airbyte Pipeline Runs	`airbyte:read`	Recent pipeline runs for a client, newest first.	`GET /client/v1/airbyte/pipelines/:clientId/runs`
`get_airbyte_pipeline_run` Get Airbyte Pipeline Run	`airbyte:read`	One pipeline run with its per-connection steps — what synced, what failed and why, and whether the completion webhook was delivered.	`GET /client/v1/airbyte/pipelines/:clientId/runs/:runId`
`update_airbyte_config` Update Airbyte Config	`airbyte:write`	Set the origins the OAuth consent flow may redirect the browser back to (replaces the stored list).	`PATCH /client/v1/airbyte/config`
`provision_airbyte` Provision Airbyte Workspace	`airbyte:write`	Provision (or adopt) this agency's Airbyte workspace.	`POST /client/v1/airbyte/config/provision`
`register_airbyte_notifications` Register Airbyte Notifications	`airbyte:write`	Point this agency's Airbyte workspace at Mythic's sync-notification receiver.	`POST /client/v1/airbyte/config/notifications`
`create_airbyte_webhook` Create Airbyte Webhook Subscription	`airbyte:write`	Subscribe an agency URL to relayed sync notifications (Mythic keeps Airbyte's single webhook slot and forwards).	`POST /client/v1/airbyte/webhooks`
`delete_airbyte_webhook` Delete Airbyte Webhook Subscription	`airbyte:write`	Remove one relay subscription.	`DELETE /client/v1/airbyte/webhooks/:id`
`create_airbyte_connection` Create Airbyte Connection	`airbyte:write`	Create a source + connection for one client.	`POST /client/v1/airbyte/connections`
`replace_airbyte_connection` Replace Airbyte Connection Settings	`airbyte:write`	PUT a connection's settings: every field you send replaces what is stored, and `streams` in particular replaces the ENTIRE enabled-stream list — a stream you leave out stops syncing, so send the complete set…	`PUT /client/v1/airbyte/connections/:id`
`delete_airbyte_connection` Delete Airbyte Connection	`airbyte:write`	Delete a connection (and its Airbyte source).	`DELETE /client/v1/airbyte/connections/:id`
`trigger_airbyte_sync` Trigger Airbyte Sync	`airbyte:write`	Start a sync job on one connection now, regardless of its schedule.	`POST /client/v1/airbyte/connections/:id/sync`
`set_airbyte_schedule` Set Airbyte Schedule	`airbyte:write`	Change one connection's own Airbyte schedule.	`PATCH /client/v1/airbyte/connections/:id/schedule`
`initiate_airbyte_oauth` Initiate Airbyte OAuth	`airbyte:write`	Start the OAuth handoff for an OAuth-capable platform.	`POST /client/v1/airbyte/oauth/initiate`
`set_airbyte_pipeline` Set Airbyte Pipeline	`airbyte:write`	Create or update a client's pipeline (orchestrated daily sync of a set of connections, driven by Mythic's cron, with a signed completion webhook once the whole set is done).	`PUT /client/v1/airbyte/pipelines/:clientId`
`start_airbyte_pipeline_run` Start Airbyte Pipeline Run	`airbyte:write`	Start a pipeline run now instead of waiting for run_hour_utc.	`POST /client/v1/airbyte/pipelines/:clientId/runs`
`airbyte_pipeline_run_action` Airbyte Pipeline Run Action	`airbyte:write`	Act on one pipeline run.	`POST /client/v1/airbyte/pipelines/:clientId/runs/:runId/retry` `POST /client/v1/airbyte/pipelines/:clientId/runs/:runId/force` `POST /client/v1/airbyte/pipelines/:clientId/runs/:runId/redeliver`

Replay Vision

AI scanners that watch session recordings with the agency’s own model key (OpenRouter, Gemini or Cloudflare): keys, scanners, on-demand scans and their observations. A key bound to one client may read that client’s scanners, observations and summaries; everything else is agency-wide.

Tool	Scope	What it does	Proxies
`get_replay_vision_stats` Get Replay Vision Stats	`replay_vision:read`	UTC inclusive range (max 90 days, default last 30) with zero-filled completion-day counts/known BYOK costs; independent current-month created_at scan-limit usage and flat per-scanner month counts including z…	`GET /client/v1/replay-vision/stats`
`list_replay_scout_templates` List Replay Scout Templates	`replay_vision:read`	Compatible read-only Scout templates and defaults, each with its sources.	`GET /client/v1/replay-vision/scout-templates`
`list_replay_scouts` List Replay Scouts	`replay_vision:read`	Scouts for one scanner, with latest_report_id, current_run, can_run_now and persisted IANA schedule.	`GET /client/v1/replay-vision/scanners/:id/scouts`
`get_replay_scout` Get Replay Scout	`replay_vision:read`	One read-only Scout configuration and lifecycle.	`GET /client/v1/replay-vision/scouts/:id`
`list_replay_scout_reports` List Replay Scout Reports	`replay_vision:read`	Frozen report summaries newest first, scoped to the current client roster.	`GET /client/v1/replay-vision/reports`
`get_replay_scout_report` Get Replay Scout Report	`replay_vision:read`	Immutable report text/sections, bounded input coverage, frozen server-built evidence, RUN-attempt cost/coverage and publish/delivery states.	`GET /client/v1/replay-vision/reports/:id`
`list_replay_vision_keys` List Replay Vision Keys	`replay_vision:read`	The agency's model keys for Replay Vision and client scouts: agency keys (scope agency) and per-client overrides (scope location), at most one per provider (openrouter, gemini, cloudflare) in each scope.	`GET /client/v1/replay-vision/keys`
`list_replay_vision_models` List Replay Vision Models	`replay_vision:read`	Models that can scan a recording (video or still frames) and return structured answers.	`GET /client/v1/replay-vision/models`
`list_replay_scanners` List Replay Scanners	`replay_vision:read`	Replay Vision scanners: saved questions an AI model answers by watching session recordings.	`GET /client/v1/replay-vision/scanners`
`get_replay_scanner` Get Replay Scanner	`replay_vision:read`	One scanner with this month's usage: queued, succeeded, failed, ineligible, how many count against monthly_scan_limit, and the model spend the provider reported (OpenRouter reports it; direct Gemini does not…	`GET /client/v1/replay-vision/scanners/:id`
`list_scanner_observations` List Scanner Observations	`replay_vision:read`	One scanner's observations.	`GET /client/v1/replay-vision/scanners/:id/observations`
`list_replay_observations` List Replay Observations	`replay_vision:read`	Agency-wide observations across the current client roster, optionally scoped to location, scanner or session.	`GET /client/v1/replay-vision/observations`
`get_replay_observation` Get Replay Observation	`replay_vision:read`	One observation with frozen scanner config/snapshot, agency-shared rating, answer, decision, provider usage and cost.	`GET /client/v1/replay-vision/observations/:id`
`get_scanner_summary` Get Scanner Summary	`replay_vision:read`	What one scanner found over a UTC date range, counted in one call: observation statuses, and per type, monitor yes/no counts with shares, classifier tag counts with shares (plus freeform tags), scorer averag…	`GET /client/v1/replay-vision/scanners/:id/summary`
`create_replay_scout` Create Replay Scout	`replay_vision:write`	Create a read-only scheduled text Scout on a scanner, max 5.	`POST /client/v1/replay-vision/scanners/:id/scouts`
`update_replay_scout` Update Replay Scout	`replay_vision:write`	Merge a Scout patch; scanner/template/sources are immutable.	`PATCH /client/v1/replay-vision/scouts/:id`
`delete_replay_scout` Delete Replay Scout	`replay_vision:write`	Soft-delete and disable a Scout, cancel pending runs; retains immutable reports/evidence.	`DELETE /client/v1/replay-vision/scouts/:id`
`run_replay_scout` Run Replay Scout	`replay_vision:write`	Queue an enabled Scout before its FIRST report only; returns 202 run_id/status pending.	`POST /client/v1/replay-vision/scouts/:id/run`
`rate_replay_observation` Rate Replay Observation	`replay_vision:write`	Set or clear an agency-shared rating on a succeeded observation only (409 otherwise).	`PUT /client/v1/replay-vision/observations/:id/rating`
`delete_replay_vision_key` Delete Replay Vision Key	`replay_vision:write`	Remove one model key: an agency key (no location_id) or one client's override (location_id), chosen by provider.	`DELETE /client/v1/replay-vision/keys`
`make_default_replay_vision_key` Set Default Replay Vision Key	`replay_vision:write`	Make one key its scope's default: the key scanners and client scouts without a provider run on.	`POST /client/v1/replay-vision/keys/default`
`create_replay_scanner` Create Replay Scanner	`replay_vision:write`	Create a scanner on one client.	`POST /client/v1/replay-vision/scanners`
`update_replay_scanner` Update Replay Scanner	`replay_vision:write`	Change a scanner.	`PATCH /client/v1/replay-vision/scanners/:id`
`delete_replay_scanner` Delete Replay Scanner	`replay_vision:write`	Delete a scanner and its observations/scouts.	`DELETE /client/v1/replay-vision/scanners/:id`
`scan_replays` Scan Replays	`replay_vision:write`	Queue up to 50 recordings (session ids from list_replays) for one scanner now.	`POST /client/v1/replay-vision/scanners/:id/scan`
`retry_replay_observation` Retry Replay Observation	`replay_vision:write`	Put a failed observation back in the queue with the same scanner snapshot.	`POST /client/v1/replay-vision/observations/:id/retry`

Scouts

Scheduled agents that read one client’s data with the agency’s own model key and file findings into the client’s inbox: scouts, runs, memory, reports and report status. Reading a report also needs the read scope of every surface the scout read.

Agency-wide keys only — a key bound to one client cannot hold this scope.

Tool	Scope	What it does	Proxies
`list_scout_skills` List Scout Skills	`scouts:read`	The built-in client scout skills with what each reads (families) and its full playbook: error_tracking (browser exceptions + server error logs that cost visitors, leads or tracking), attribution_drift (paid…	`GET /client/v1/scouts/skills`
`list_scout_models` List Scout Models	`scouts:read`	Tool-capable models a scout can run on, for one provider (the client's or agency's model key).	`GET /client/v1/scouts/models`
`get_scout_usage` Get Scout Usage	`scouts:read`	Reported BYOK spend on scout model calls: runs, calls, known cost and calls whose cost the provider did not report, per scout.	`GET /client/v1/scouts/usage`
`list_scout_reports` List Scout Reports	`scouts:read`	The inbox: scout reports newest first, one finding each, with title, headline, priority (P0-P4), actionability, status and the scout that filed it.	`GET /client/v1/scouts/reports`
`get_scout_report` Get Scout Report	`scouts:read`	One report: summary, actionability and its explanation, suggested next step, status history, and evidence.	`GET /client/v1/scouts/reports/:id`
`list_scouts` List Scouts	`scouts:read`	Client scouts (agency-wide, or one client with location_id): skill, schedule, model, enabled, next/last run, last error and any run in progress.	`GET /client/v1/scouts`
`get_scout` Get Scout	`scouts:read`	One client scout's configuration and state.	`GET /client/v1/scouts/:id`
`list_scout_runs` List Scout Runs	`scouts:read`	A scout's last 30 runs: status, its close-out summary (what it looked at, filed, ruled out), turns, tool calls, reports filed/edited, surfaces read, model, and reported spend.	`GET /client/v1/scouts/:id/runs`
`get_scout_memory` Get Scout Memory	`scouts:read`	What a scout remembers between runs: pattern: (baselines), noise: (what it skips), dedupe:, addressed: (fixed), report: (which report tracks what).	`GET /client/v1/scouts/:id/memory`
`update_scout_report` Update Scout Report	`scouts:write`	Triage a report: resolve (terminal; a recurrence is filed as a new report), dismiss with a reason (the scout reads it on its next runs, for 30 days, so it stops re-filing), snooze until a time (max 90 days),…	`PATCH /client/v1/scouts/reports/:id`
`create_scout` Create Scout	`scouts:write`	Create a scheduled client scout that reads the client's data with the agency's own model key and files one report per finding into the inbox.	`POST /client/v1/scouts`
`update_scout` Update Scout	`scouts:write`	Change a scout's name, instructions, schedule, delivery, model, or pause it (enabled:false).	`PATCH /client/v1/scouts/:id`
`delete_scout` Delete Scout	`scouts:write`	Delete a scout.	`DELETE /client/v1/scouts/:id`
`run_scout` Run Scout Now	`scouts:write`	Queue a run now (the scout must be enabled with no run in progress).	`POST /client/v1/scouts/:id/run`
`forget_scout_memory` Forget Scout Memory	`scouts:write`	Delete one memory entry by key (for example a noise: entry that hides something that now matters).	`DELETE /client/v1/scouts/:id/memory/:key`

Alerts

The central alert feed across every client, the rules that route alerts, and the Slack/webhook channels they are delivered to.

Agency-wide keys only — a key bound to one client cannot hold this scope.

Tool	Scope	What it does	Proxies
`list_alerts` List Alerts	`alerts:read`	The central alert feed across every client, newest first.	`GET /client/v1/alerts/events`
`get_alert` Get Alert	`alerts:read`	One alert with its payload and every delivery attempt (which channel, which rule routed it, ok or the error).	`GET /client/v1/alerts/events/:id`
`list_alert_rules` List Alert Rules	`alerts:read`	The agency's routing rules: which sources, for which clients (empty = all, including future clients), at or above which severity, go to which channels.	`GET /client/v1/alerts/rules`
`list_alert_channels` List Alert Channels	`alerts:read`	Where alerts are delivered: Slack incoming webhooks and signed HTTPS webhooks.	`GET /client/v1/alerts/channels`
`resolve_alert` Resolve Alert	`alerts:write`	Mark an open alert resolved by hand.	`PATCH /client/v1/alerts/events/:id`
`create_alert_rule` Create Alert Rule	`alerts:write`	Route alerts to channels.	`POST /client/v1/alerts/rules`
`update_alert_rule` Update Alert Rule	`alerts:write`	Change any field of a rule.	`PATCH /client/v1/alerts/rules/:id`
`delete_alert_rule` Delete Alert Rule	`alerts:write`	Delete a rule.	`DELETE /client/v1/alerts/rules/:id`
`create_alert_channel` Create Alert Channel	`alerts:write`	Add a delivery target.	`POST /client/v1/alerts/channels`
`update_alert_channel` Update Alert Channel	`alerts:write`	Rename a channel or turn it on/off.	`PATCH /client/v1/alerts/channels/:id`
`delete_alert_channel` Delete Alert Channel	`alerts:write`	Delete a channel.	`DELETE /client/v1/alerts/channels/:id`
`test_alert_channel` Test Alert Channel	`alerts:write`	Send a test message to one channel now and report whether it was accepted (HTTP status or error).	`POST /client/v1/alerts/channels/:id/test`

Agency

The client roster, cross-location stats, usage, quota, destination health and contract summaries, and agency settings such as OAuth return origins.

Agency-wide keys only — a key bound to one client cannot hold this scope.

Tool	Scope	What it does	Proxies
`get_agency_stats` Get Agency Stats	`agency:read`	Cross-location rollup for the whole agency: one row per location with event volume, unique users, unique sessions over a date window, plus the all-time live person count.	`GET /client/v1/agency/stats`
`get_agency_destination_health` Get Agency Destination Health	`agency:read`	Delivery health of every location's activation destinations over a trailing window — one row per (location, destination) with success rate, failure counts, and the last error.	`GET /client/v1/agency/destinations/health`
`get_agency_destination_quality` Get Agency Destination Quality	`agency:read`	Meta dataset quality for every client with a Meta CAPI destination, worst first — the answer to 'which of my clients has bad Meta tracking'.	`GET /client/v1/agency/destinations/quality`
`get_agency_destination_quality_trend` Get Agency Destination Quality Trend	`agency:read`	How every client's Meta match quality MOVED, day by day — one row per (day, location) with that day's worst scored event, its match quality, and the critical/warning counts.	`GET /client/v1/agency/destinations/quality/trend`
`get_agency_last_event` Get Agency Last Event	`agency:read`	Tracking heartbeat for every client in one call: events received and the latest event time per client over hours_back (default 24).	`GET /client/v1/agency/last-event`
`list_agency_events` List Agency Events	`agency:read`	The newest raw events across every client in one call, newest first: event name, client (location_id, location_name), person ids, session position, page URL, UTM, device, country and the full properties JSON.	`GET /client/v1/agency/events`
`get_agency_event_types` Get Agency Event Types	`agency:read`	Every event name seen across the roster in hours_back, with its total count summed over all clients, largest first.	`GET /client/v1/agency/events/summary`
`get_agency_error_summary` Get Agency Error Summary	`agency:read`	JavaScript error tracking for every client in one call, most exceptions first: exceptions in [date_from, date_to] and the equal window before it, sessions with an exception, all sessions, crash-free session…	`GET /client/v1/agency/exceptions/summary`
`get_agency_link_uptime` Get Agency Link Uptime	`agency:read`	Short link uptime for every client in one call: each link checked in the last 24 hours with its client, title, destination, whether the latest check was up, status code or error, response time, 24h uptime, a…	`GET /client/v1/agency/links/uptime`
`get_agency_contracts_summary` Get Agency Contracts Summary	`agency:read`	Per location: how many event contracts are declared, whether the daily contract-alert webhook is configured, whether an agency alert rule routes contract alerts for it (alert_rule_routed), and when the last…	`GET /client/v1/agency/contracts/summary`
`get_agency_contract_status` Get Agency Contract Status	`agency:read`	Yesterday's (UTC) event contract result for every client in one call: status ('ok', 'violations', 'no_data', 'no_contracts', 'not_in_plan' or 'error'), contracts checked, how many events violated or had no d…	`GET /client/v1/agency/contracts/status`
`get_agency_usage` Get Agency API Usage	`agency:read`	The agency's own API + MCP usage, daily per location/surface/route (or per surface/route with group_by 'route', which fits 30 days at any volume): request counts, auth failures, rate-limited and error counts…	`GET /client/v1/agency/usage`
`get_agency_quota` Get Agency Plan & Event Quota	`agency:read`	The agency's plan, the features it includes, and month-to-date ingested events against the quota.	`GET /client/v1/agency/quota`
`get_agency_oauth_return_origins` Get OAuth Return Origins	`agency:read`	The origins (scheme://host[:port]) a browser may be sent back to after a GoHighLevel install or a Google destination consent, via their return_url.	`GET /client/v1/agency/oauth-return-origins`
`refresh_agency_destination_quality` Refresh Agency Destination Quality	`agency:write`	Re-read Meta's match quality now for every client in the agency (or only location_ids) and overwrite today's snapshot, so get_agency_destination_quality matches Events Manager instead of the 07:00 UTC capture.	`POST /client/v1/agency/destinations/quality/refresh`
`set_agency_oauth_return_origins` Set OAuth Return Origins	`agency:write`	Replace the agency's OAuth return-origin list (the whole list, not an append).	`PUT /client/v1/agency/oauth-return-origins`

Ads Manager extension

Read-only bootstrap for the Mythic browser extension: agency roster with ad-platform flags and ad-account to client resolution.

Agency-wide keys only — a key bound to one client cannot hold this scope.

Tool	Scope	What it does	Proxies
`resolve_ad_account` Resolve Ad Account to Client	`extension:read`	Which of the agency's clients syncs a given Meta or Google ad account, from the clients' Airbyte ad-platform connections.	`GET /client/v1/extension/resolve`

Was this page helpful?

Tool Reference

About this reference

Session

People

Events & Explore

Sessions

Session Replays

Error Tracking

Logs

Heatmaps

BigQuery Export

Insights

Dashboards

Canvases

Segments

Constraint Finder

Correlation

Incrementality

Event Contracts

Destinations

Ad UTM Auto-tagging

Tracking Settings

Client Provisioning

Client Setup

Links & Uptime

Tags

Snippets

Transformers

Airbyte Sync

Replay Vision

Scouts

Alerts

Agency

Ads Manager extension