InsideOut (Riley)

{
  "id": "aix_364ec2289b",
  "name": "InsideOut (Riley)",
  "operator": "(unclaimed - source: registry-official · publisher: com.luthersystems.insideout)",
  "depth": 2,
  "status": "unclaimed",
  "last_crawled": "2026-09-25",
  "missing_fields": [
    "pricing",
    "operator.identity"
  ],
  "skills": [
    "cache-store",
    "cloud-ops",
    "sql-database"
  ],
  "protocols": {
    "mcp": "https://app.luthersystems.com/v1/insideout-mcp",
    "a2a": null
  },
  "pricing": null,
  "regions": [
    "global"
  ],
  "languages": [
    "en"
  ],
  "reputation": {
    "tasks_completed": null,
    "dispute_rate": null,
    "p95_latency_ms": 3298,
    "uptime_30d": 1,
    "onchain_volume_30d_usd": null
  },
  "aix_score": 41,
  "verification": {
    "identity": "none",
    "health": "probe/24h",
    "pricing": "unknown",
    "last_check": "2026-09-25T21:01:06.988Z"
  },
  "pricing_model": "unknown",
  "links": [
    {
      "label": "homepage",
      "url": "https://insideout.luthersystems.com/"
    },
    {
      "label": "repository",
      "url": "https://github.com/luthersystems/insideout-agent-skills"
    }
  ],
  "avatar": "https://github.com/luthersystems.png?size=160",
  "socials": [
    {
      "label": "github",
      "url": "https://github.com/luthersystems"
    }
  ],
  "profile": {
    "mcp_server": "insideout-agent",
    "mcp_version": "v2.0.0",
    "tool_count": 24,
    "tools": [
      {
        "name": "awsinspect",
        "description": "INSPECTION: Inspect AWS infrastructure for a deployed project\n⚠️ **PREREQUISITE**: This tool requires a prior deployment ATTEMPT (successful or failed).\nCheck convostatus for hasDeployAttempt=true before calling. Works even after failed deploys to inspect orphaned resources.\n\nInspect deployed AWS resources after a deployment attempt.\nUse this tool when the user asks about the status or details of their deployed infrastructure.\nIt fetches temporary read-only credentials securely and queries the AWS API directly.\n\nRESPONSE TIERS (default is summary for token efficiency):\n- Summary (default): Key"
      },
      {
        "name": "awsinspect_batch",
        "description": "BATCH INSPECTION: run up to 32 AWS inspect probes in one call.\n⚠️ **PREREQUISITE**: Same as awsinspect — deploy attempt required.\nCheck convostatus for hasDeployAttempt=true before calling.\n\nUse this when you need to check more than ~3 resources. The backend fetches\nOracle credentials ONCE per batch and fans out probes against a single AWS\nconfig — for a 12-resource health check this is ~5–8× faster and 12× fewer\nOracle round-trips than calling awsinspect 12 times.\n\nBUDGETS:\n- Up to 32 sub-probes per call (subs array length).\n- 30s per-sub timeout; 60s total batch wall-clock.\n- Concurrency cap"
      },
      {
        "name": "convoawait",
        "description": "Wait for a pending response from Riley after a convoreply timeout.\n\n🎯 USE THIS TOOL WHEN: convoreply returned a timeout error.\nThis allows you to continue waiting for the response without resending the message.\n\nREQUIRES:\n- session_id: from convoopen response\n\nOPTIONAL:\n- message_id: if known (from convoreply timeout error)\n- timeout (integer): seconds to wait. For Cursor, use 50 (default). Max 55.\n\nReturns the same format as convoreply when successful."
      },
      {
        "name": "convoinspect",
        "description": "INSPECTION: View a session's conversation transcript and metadata\nReturns the full message history (user / assistant / tool turns) plus the session's meta — workflow step, cloud, deployment status, drift state.\n\nThis is the transcript-reader companion to the other read tools — combine it with:\n  • `convostatus` for the live stack / config / pricing\n  • `tfruns` for deployment history (apply / destroy / plan / drift)\n  • `stackversions` for the stack-version ladder\n\nUse it when a user asks 'what did I say earlier?' or you need to retrace why the session ended up where it did. Read-only; never m"
      },
      {
        "name": "convoopen",
        "description": "WORKFLOW: Step 1 of 4 - Start infrastructure design conversation\nOpen an InsideOut V2 session and receive the assistant's intro message.\nThe response contains a clean message from Riley (the infrastructure advisor) - display it to the user.\n⚠️ Riley will ask questions - forward these to the user, DO NOT answer on their behalf.\nCRITICAL: This tool returns a session_id in the response metadata. You MUST use this session_id for ALL subsequent tool calls (convoreply, tfgenerate, tfdeploy, etc.).\n⚠️ The session_id includes a ?token=... suffix (format: sess_v2_xxx?token=yyy) which is part of the ses"
      },
      {
        "name": "convoreply",
        "description": "WORKFLOW: Step 2 of 4 - Continue infrastructure design conversation\nSend a user message to the active InsideOut session and receive the assistant reply.\nThe response contains a clean message from Riley - display it to the user.\n\n⚠️ CRITICAL: DO NOT answer Riley's questions yourself! Forward questions to the user and wait for their response. NEVER fabricate or assume the user's answer, even if you think you know what they would say.\nExamples of questions Riley asks that YOU MUST forward to the user:\n- 'Any questions or tweaks to these details?'\n- 'Ready for the cost estimate?'\n- 'Do you want to"
      },
      {
        "name": "convostatus",
        "description": "INSPECTION: View the current infrastructure stack for a session\nReturns the current state of the user's infrastructure design including:\n\n**Components** - Selected infrastructure services (VPC, databases, caching, etc.)\n  • Shows what services the user has chosen (e.g., PostgreSQL, Redis, S3)\n  • Includes architecture decisions (EKS vs EC2, monolith vs microservices)\n\n**Config** - Configuration details for each component\n  • Database sizes, replica counts, storage amounts\n  • Cache settings, queue configurations\n  • Backup schedules and retention policies\n\n**Pricing** - Cost estimates (when av"
      },
      {
        "name": "credawait",
        "description": "Wait for the user to securely connect their cloud account and subscribe to Luther Systems.\nPolls until credentials appear on the session.\n\n🎯 USE THIS TOOL WHEN: tfdeploy returns an 'auth_required', 'no_credentials', or 'credentials_expired' error.\n\nThe user needs to visit the connect URL to:\n1. Connect their cloud credentials (AWS or GCP)\n2. Sign up and subscribe to a Luther Systems plan (required for deployment)\n\nThis secure connection allows InsideOut to deploy and manage infrastructure in the user's cloud account on their behalf. Credentials are handled securely and only used for deploymen"
      },
      {
        "name": "gcpinspect",
        "description": "INSPECTION: Inspect GCP infrastructure for a deployed project\n⚠️ **PREREQUISITE**: This tool requires a prior deployment ATTEMPT (successful or failed).\nCheck convostatus for hasDeployAttempt=true before calling. Works even after failed deploys to inspect orphaned resources.\n\nInspect deployed GCP resources after a deployment attempt.\nUse this tool when the user asks about the status or details of their deployed GCP infrastructure.\nIt fetches temporary read-only credentials securely and queries the GCP API directly.\n\nRESPONSE TIERS (default is summary for token efficiency):\n- Summary (default):"
      },
      {
        "name": "gcpinspect_batch",
        "description": "BATCH INSPECTION: run up to 32 GCP inspect probes in one call.\n⚠️ **PREREQUISITE**: Same as gcpinspect — deploy attempt required.\nCheck convostatus for hasDeployAttempt=true before calling.\n\nUse this when you need to check more than ~3 resources. The backend fetches\nOracle credentials ONCE per batch and fans out probes against a single GCP\ncredentials blob — a 12-resource health check is ~5–8× faster and 12× fewer\nOracle round-trips than calling gcpinspect 12 times.\n\nBUDGETS:\n- Up to 32 sub-probes per call (subs array length).\n- 30s per-sub timeout; 60s total batch wall-clock.\n- Concurrency ca"
      },
      {
        "name": "help",
        "description": "Get workflow guidance for using InsideOut infrastructure tools.\nCall help() for a compact overview, or help(section=...) for a detailed guide.\nSections: workflow, tools, examples, inspect.\nResponses include hints with next_actions and related_tools."
      },
      {
        "name": "stackdiff",
        "description": "Structured diff showing what would be deployed if the user ran tfdeploy now.\nReturns component-level changes (added/removed/modified), field-level details, and pricing deltas.\n\nDefaults (#1392): with no version arguments, compares the LAST SUCCESSFULLY DEPLOYED version against the user's CURRENT LIVE DESIGN (the same data the UI shows). Empty baseline if nothing has been deployed or after a destroy. Pending drafts are NOT used as the target — they go stale once the user edits past them; live IR via chat history is always current.\n\nPass explicit `from_version` and/or `to_version` integers to co"
      },
      {
        "name": "stackrollback",
        "description": "Create a draft version by reverting to a previous version's config.\nCopies components, config, and pricing from the target version. If a draft already exists, updates it in-place (single-draft rule).\n\nUse `stackversions` first to find available version numbers.\n\nREQUIRES: session_id from convoopen response (format: sess_v2_...), version (target version number)."
      },
      {
        "name": "stackversions",
        "description": "List all stack versions for a session (newest first).\nShows version history including version number, status (draft/confirmed/applied), change summaries, and timestamps.\n\nUse this tool to see the design history, review what changed between iterations, or find a version number to roll back to.\n\nREQUIRES: session_id from convoopen response (format: sess_v2_...)."
      },
      {
        "name": "submit_feedback",
        "description": "FEEDBACK: Submit feedback, bug reports, or feature requests to Luther Systems\nUse this tool to forward user feedback directly to the Luther Systems team. This includes bug reports, feature requests, questions, or general feedback about InsideOut.\nThe agent itself can also use this tool to report issues it encounters during operation.\n\nREQUIRES: session_id, category, message\nOPTIONAL: user_email (for follow-up), user_name, source (default: 'mcp'), initiator ('user' or 'agent')\n\nCategories: bug_report, feature_request, general_feedback, question, security\n\nThe 'initiator' field tracks who trigge"
      },
      {
        "name": "tfdeploy",
        "description": "WORKFLOW: Step 4 of 4 - Deploy infrastructure to the cloud\nDeploy infrastructure by starting a Terraform job for an InsideOut session.\nThis tool initiates the actual deployment process after Terraform files have been generated.\nIMPORTANT: This starts a long-running job (15+ minutes). Use tfstatus to monitor progress.\nSINGLE-FLIGHT: only one TF job (apply/plan/destroy/drift) runs per session at a time. If another job is already in flight, tfdeploy returns tf_job_conflict with the live job_id — attach with tfstatus/tflogs instead of retrying, or pass force_new=true to override.\nReturns confirmat"
      },
      {
        "name": "tfdestroy",
        "description": "DESTROY: Tear down previously deployed infrastructure\nDestroys infrastructure by calling the Oracle destroy endpoint for a session that has a prior successful deployment.\nIMPORTANT: This starts a long-running job. Use tfstatus/tflogs to monitor progress.\nSINGLE-FLIGHT: only one TF job per session at a time. If another job is already in flight, tfdestroy returns tf_job_conflict with the live job_id — attach with tfstatus/tflogs, or pass force_new=true to override.\nREQUIRES: session_id from convoopen response (format: sess_v2_...).\nOPTIONAL: force_new (boolean, default false) - bypass the single"
      },
      {
        "name": "tfdrift",
        "description": "DRIFT CHECK: Run a read-only drift detection check\nChecks whether deployed infrastructure has drifted from the expected Terraform state.\nThis is a read-only operation — it does NOT modify any infrastructure.\nReturns job_id. Use tflogs to stream the drift check results.\nSINGLE-FLIGHT: only one TF job per session at a time. If another job is already in flight, tfdrift returns tf_job_conflict with the live job_id — attach with tfstatus/tflogs, or pass force_new=true to override.\nREQUIRES: session_id from convoopen response (format: sess_v2_...).\nPREREQUISITE: The session must have a prior deploym"
      },
      {
        "name": "tfgenerate",
        "description": "WORKFLOW: Step 3 of 4 - Generate Terraform files from completed design\nGenerate Terraform files from an InsideOut session that has completed infrastructure design.\n\n⚠️ PREREQUISITE: Only call this AFTER convoreply returns with `terraform_ready=true` in the response metadata.\nDO NOT call this while convoreply is still running or before terraform_ready is confirmed!\nIf you get 'session has not reached terraform-ready state', wait for convoreply to complete first.\n\n🎯 USE THIS TOOL WHEN: convoreply has returned with terraform_ready=true, OR the user asks to 'see the terraforms', 'generate terrafo"
      },
      {
        "name": "tflogs",
        "description": "MONITORING: Fetch Terraform deployment logs with pagination\nFetches logs from a running or completed Terraform deployment job.\nFor **completed jobs**: uses REST endpoint for instant retrieval (supports `tail` for server-side filtering).\nFor **running jobs**: streams via SSE with timeout-based pagination.\n\n**PAGINATION** (running jobs only): Use `last_event_id` from the response to fetch more:\n1. First call: `tflogs(session_id='...')` → get logs + `last_event_id`\n2. Next call: `tflogs(session_id='...', last_event_id='...')` → get NEW logs only\n3. Repeat until `complete: true` in response\n\n**RES"
      },
      {
        "name": "tfoutputs",
        "description": "INSPECTION: Retrieve Terraform outputs from a completed deployment\nReturns structured output values (VPC IDs, endpoints, cluster names, etc.) after a successful deploy.\nSensitive outputs are redacted (shown as '(sensitive)').\n\nBy default returns outputs for the latest successful deploy. Optionally specify job_id to get outputs for a specific deployment.\n\nREQUIRES: session_id from convoopen response (format: sess_v2_...).\nOPTIONAL: job_id (specific deployment), lifecycle (filter by step e.g. 'cloud-provision')."
      },
      {
        "name": "tfplan",
        "description": "PREVIEW: Run terraform plan to preview infrastructure changes\nRuns a terraform plan for an InsideOut session without applying any changes.\nThis lets the user review what will be created/changed/destroyed before committing.\nReturns job_id, plan_id, and project_id. Use tflogs to stream the plan output.\nAfter the plan completes, use tfdeploy with plan_id to apply the exact plan.\nSINGLE-FLIGHT: only one TF job per session at a time. If another job is already in flight, tfplan returns tf_job_conflict with the live job_id — attach with tfstatus/tflogs, or pass force_new=true to override.\nREQUIRES: s"
      },
      {
        "name": "tfruns",
        "description": "INSPECTION: List all Terraform deployment runs for a session\nReturns job IDs, statuses, types (apply/destroy), and timestamps for every run.\nUse this to see deployment history, find job IDs for log inspection, or check which deployments succeeded or failed.\n\nREQUIRES: session_id from convoopen response (format: sess_v2_...)."
      },
      {
        "name": "tfstatus",
        "description": "MONITORING: Quick status check for Terraform deployments\nCheck the current status of a Terraform deployment job.\nUse this tool to quickly check if a deployment is running, completed, or failed.\nReturns job status, job_id, and other metadata without streaming logs.\nUse tflogs to stream the actual deployment logs.\nREQUIRES: session_id from convoopen response (format: sess_v2_...).\nOPTIONAL: job_id to target a specific deployment (use tfruns to discover IDs).\n\n**LIVENESS**: The response carries two distinct timestamps:\n- `updated_at` — last semantic change (only bumped when status / drift / versi"
      }
    ],
    "profiled_at": "2026-09-25T21:01:06.988Z"
  },
  "unreachable": false
}