Sleepwalker
Server Details
AI Visibility and Content Intelligence tools for Claude and MCP-compatible agents.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- OAuth
- Works in Glama
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- Repository
- followanton/sleepwalker
- GitHub Stars
- 22
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 19 tools
Seven retrieval tools (get_summaries, get_test_results, get_run_result, content_run_status, visibility_run_status, prompt_response, reports_by_url) overlap in purpose, but descriptions explicitly cross-reference each other and clarify scoping (saved-test vs direct-URL, compact vs full, single-answer vs full-run). The create/discover/score/suggest tools are clearly distinct. An agent can route correctly with care, but the retrieval surface is dense enough to risk misselection.
Every tool follows a strict verb_noun pattern with a uniform 'sleepwalker' namespace infix (create_sleepwalker_visibility_run, get_sleepwalker_prompt_response, list_sleepwalker_tests). Verbs are drawn from a small predictable set (cancel, create, discover, get, list, score, serialize, suggest) and applied consistently. No stylistic deviations.
19 tools for a platform spanning Content Intelligence, AI Visibility, and AI Citations (across saved-test and direct-URL run modes) is justified rather than padded. It sits at the high end of the comfortable range, with the status/list/get clusters accounting for much of the count, but each maps to a real resource/action.
The surface covers the run lifecycle (create, list, status, result, cancel) plus discovery, scoring, serialization, and prompt suggestion for both major workflows. Gaps exist: only visibility runs have a cancel tool (no equivalent for content runs) and there is no delete/update for saved tests. These are minor and workaroundable for a run-oriented system.
Available Tools
19 toolscancel_sleepwalker_visibility_runCancel Sleepwalker visibility runADestructiveIdempotentInspect
Cancel the queued remainder of an AI Visibility run (the killswitch).
Use when the user wants to stop a direct-URL run (created by create_sleepwalker_visibility_run) that is queued or running. Probes no worker has picked up yet never run and their reserved credits return to the account; probes a worker already claimed (they are dispatched in batches) finish and settle normally, and completed results stay available. While work is in flight the response reports billing_pending true and the remaining release normally happens when the last probe finishes; if a run later looks stuck (still running with nothing progressing, for example after a worker crash), call this tool again to force the remainder out and release the credits. Idempotent: cancelling a finished run, or cancelling twice, changes nothing and reports the current state. Poll get_sleepwalker_visibility_run_status for final counts; released credits equal reserved minus settled once terminal.
Inputs: run_id (required). Returns: object
create_sleepwalker_content_runCreate Sleepwalker Content Intelligence runAInspect
Queue a persisted Content Intelligence run.
Send url for a new direct-URL run, or test_id to rerun an existing saved CI test. Use only after explicit user confirmation. This starts queued work and returns quickly with a run_id; poll get_sleepwalker_content_run_status until the run is terminal instead of creating duplicate runs. The response includes billing fields when credits are reserved.
Inputs: url, country, test_id, language, +2 more. Returns: object
create_sleepwalker_visibility_runCreate Sleepwalker visibility runAInspect
Queue a saved AI Visibility run from prompts x platforms or explicit probes.
Use this when the user wants reportable probe results, not just prompt ideas. This starts queued work and returns quickly with a run_id; poll get_sleepwalker_visibility_run_status until the run is terminal instead of creating duplicate runs. Platforms accept canonical slugs or common labels: perplexity, openai/ChatGPT, grok, gemini. To pin AI models, pass the models map with matrix mode (per platform) or set model on explicit probes (per probe); values are model ids from list_sleepwalker_visibility_models or the keywords latest / prior / default. Every model has its own per-probe price, defaults included; call list_sleepwalker_visibility_models for the exact credits_per_probe before quoting cost. Results record the exact model that answered each probe. The response includes credit fields when credits are reserved.
Inputs: url (required), target_entity (required), models, probes, +6 more. Returns: object
discover_sleepwalker_content_trendsDiscover Sleepwalker content trendsAInspect
Find relevant content trends for a URL. This runs extraction and serialization internally and returns compact industry, trend, and citation data only.
Inputs: url (required), country, language, extraction_mode. Returns: object
get_sleepwalker_consistent_recommendationsGet Sleepwalker consistent recommendationsARead-onlyIdempotentInspect
Find recurring Content Intelligence recommendations across recent runs for a known test.
Inputs: test_id (required), runs, min_frequency. Returns: object
get_sleepwalker_content_run_statusGet Sleepwalker Content Intelligence run statusARead-onlyIdempotentInspect
Read status and optional full result for a persisted direct-URL Content Intelligence run.
Poll compact status while queued or running. Set include_result only after completion or when the user explicitly needs the full result payload.
Inputs: run_id (required), include_result. Returns: object
get_sleepwalker_prompt_responseGet Sleepwalker prompt responseARead-onlyIdempotentInspect
Fetch one capped AI Visibility prompt response from any run, saved-test or direct-URL. Use result_index or prompt_index for the flattened result list (same order as the status ca_result), or platform + prompt_index for the Nth prompt on a platform. Preferred over full include_results for reading large runs.
Inputs: run_id (required), offset, platform, max_chars, +3 more. Returns: object
get_sleepwalker_reports_by_urlGet Sleepwalker reports by URLARead-onlyIdempotentInspect
Find owned CI/AC tests by URL and fetch their latest UI-visible runs.
Inputs: url (required), days, limit, test_type. Returns: object
get_sleepwalker_run_resultGet Sleepwalker run resultARead-onlyIdempotentInspect
Fetch one full structured CI/AC run belonging to a saved test. Direct-URL runs created by create_sleepwalker_content_run or create_sleepwalker_visibility_run are read with their run-status tools instead. AC raw responses are fetched separately with get_sleepwalker_prompt_response.
Inputs: run_id (required). Returns: object
get_sleepwalker_summariesGet Sleepwalker summariesARead-onlyIdempotentInspect
Scan compact CI/AC run summaries from the last 1-30 days before drilling into full runs.
Inputs: days, limit, test_id, test_type, +1 more. Returns: object
get_sleepwalker_test_resultsGet Sleepwalker test resultsARead-onlyIdempotentInspect
Fetch compact latest run history for a known test. Use get_sleepwalker_run_result for full details.
Inputs: test_id (required), limit. Returns: object
get_sleepwalker_visibility_run_statusGet Sleepwalker visibility run statusARead-onlyIdempotentInspect
Read status, probe progress, and optional results for a queued infrastructure AI Visibility run.
Poll compact status while queued or running. Set include_results only after completion or when the user explicitly needs full response and citation data. For large runs, page with results_offset/results_limit (ca_result reports results_total), or fetch one answer at a time with get_sleepwalker_prompt_response; the summary always covers the full run.
Inputs: run_id (required), results_limit, include_probes, results_offset, +1 more. Returns: object
list_sleepwalker_content_runsList Sleepwalker Content Intelligence runsARead-onlyIdempotentInspect
List persisted direct-URL Content Intelligence runs created through Sleepwalker infrastructure actions. When has_more is true, pass next_starting_after as starting_after to fetch the next page.
Inputs: limit, status, starting_after. Returns: object
list_sleepwalker_testsList Sleepwalker testsARead-onlyIdempotentInspect
List the authenticated user's Content Intelligence and AI Citations tests only.
Inputs: limit, test_type, brand_name. Returns: object
list_sleepwalker_visibility_modelsList Sleepwalker visibility modelsARead-onlyIdempotentInspect
List the AI models selectable for visibility runs, per platform, with prices.
Returns every platform's options (id, label, default flag, tier, credits per probe) plus a roles table mapping default / latest / prior to a concrete model id for every platform. Use the ids or role keywords in the models map or in probes[].model when creating a run. Read-only and free.
Inputs: no named parameters. Returns: object
list_sleepwalker_visibility_runsList Sleepwalker visibility runsARead-onlyIdempotentInspect
List queued infrastructure AI Visibility runs for the authenticated user. When has_more is true, pass next_starting_after as starting_after to fetch the next page.
Inputs: limit, status, starting_after. Returns: object
score_sleepwalker_contentScore Sleepwalker contentAInspect
Score a URL with Content Intelligence without saving a run. Prefer passing compact trends from discover_sleepwalker_content_trends; omitting trends makes Sleepwalker discover them first and can be slower.
Inputs: url (required), trends, country, industry, +2 more. Returns: object
serialize_sleepwalker_page_contentSerialize Sleepwalker page contentAInspect
Use this only when you need Sleepwalker's normalized page content view. It does not score content, discover trends, suggest prompts, or save a run.
Inputs: url (required), offset, max_chars, extraction_mode. Returns: object
suggest_sleepwalker_visibility_promptsSuggest Sleepwalker visibility promptsAInspect
Suggest AI Visibility prompts for a URL. Use this for prompt ideas only; it does not query AI platforms or save a visibility run.
Inputs: url (required), brand_name (required), country, language. Returns: object
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
AI visibility & recommendation monitoring for ChatGPT, Claude, Gemini & Perplexity.
AI-visibility monitoring for your brand across ChatGPT, Claude, Perplexity & Gemini.
100+ MCP tools for AI agents: content metadata, trade intelligence, business-expertise analysis.
Build and manage AI-native customer support agents from Claude or any MCP client.
Related MCP Servers
- FlicenseNot gradedqualityCmaintenanceProvides real-time monitoring of AI agents, context, usage limits, workflows, files, Git, tests, builds, errors, secrets, and model-economy advice for tools like Claude Code, Codex, and Cursor, with 30 MCP tools for comprehensive observability.1-
- AlicenseNot gradedqualityCmaintenanceEnables auditing and monitoring brand visibility, citations, and competitor share-of-voice across ChatGPT, Perplexity, Claude, Gemini, and DeepSeek from MCP-compatible clients such as Cursor and Claude Desktop.9 npm35MIT
- AlicenseAqualityDmaintenanceProvides prompt injection detection, PII/secrets redaction, and an audit trail for AI agents via MCP tools.4MIT

flexorch-mcpofficial
AlicenseAqualityBmaintenanceEnables Claude and other MCP-compatible agents to process documents, extract structured data, detect PII, and export LLM-ready datasets through natural language tool calls.871 PyPI1MIT
Glama MCP Gateway
Add one secure layer between your agents and this server.