Liminal

Official27 toolsLive

by jaybro2042-alt · Dockerfile

Augment your work with Liminal. Stop repeating yourself and build on what’s already solved.

ai.physea/liminality MCP Server

The ai.physea/liminality MCP server (Liminality) breaks a hard question or decision into sub-questions, ties each sub-question to a real tool, and returns a worked answer you can check. It is intended for grounding complex requests in tool-backed steps rather than unverified reasoning.

🛠️ Key Features

Decomposes a hard question or decision into sub-questions
Grounds each sub-question to a real tool
Returns a worked answer that can be checked
Labeled as an MCP server in the provided description

🚀 Use Cases

Tough questions
Real decisions
Multi-step tasks that benefit from sub-question decomposition

⚡ Developer Benefits

Produces tool-grounded sub-questions for verification
Helps structure multi-step reasoning into checkable parts

⚠️ Limitations

The provided material does not specify tool coverage, supported inputs, or failure modes

problem decompositiondecision makingstructured reasoningsub-questionssolution planning

Topics

ai-agentsdecision-supportmcpmcp-servermodel-context-protocol

Captured live from the server via tools/list.

solve

Decompose, ground, and answer a non-trivial task. Supports saved or per-call depth, safe trace, caller tools, and lean/full output.

Parameters12

question	string	required
context	string	optional
resources	string	optional
tool_manifest	any	optional
depth	any	optional
max_reduction_depth	any	optional
trace	any	optional
user_id	any	optional
verbose	boolean	optional
preference_policy	any	optional
context_brief_id	string	optional
follow_up_of	string	optional

Raw schema

{
  "type": "object",
  "properties": {
    "question": {
      "title": "Question",
      "type": "string"
    },
    "context": {
      "default": "",
      "title": "Context",
      "type": "string"
    },
    "resources": {
      "default": "",
      "title": "Resources",
      "type": "string"
    },
    "tool_manifest": {
      "anyOf": [
        {
          "items": {
            "additionalProperties": true,
            "type": "object"
          },
          "type": "array"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Tool Manifest"
    },
    "depth": {
      "anyOf": [
        {
          "enum": [
            "shallow",
            "standard",
            "deep"
          ],
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Depth"
    },
    "max_reduction_depth": {
      "anyOf": [
        {
          "type": "integer"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Max Reduction Depth"
    },
    "trace": {
      "anyOf": [
        {
          "enum": [
            "off",
            "summary",
            "verbose"
          ],
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Trace"
    },
    "user_id": {
      "anyOf": [
        {
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "User Id"
    },
    "verbose": {
      "default": false,
      "title": "Verbose",
      "type": "boolean"
    },
    "preference_policy": {
      "anyOf": [
        {
          "additionalProperties": true,
          "type": "object"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Preference Policy"
    },
    "context_brief_id": {
      "default": "",
      "title": "Context Brief Id",
      "type": "string"
    },
    "follow_up_of": {
      "default": "",
      "title": "Follow Up Of",
      "type": "string"
    }
  },
  "required": [
    "question"
  ],
  "title": "solveArguments"
}

get_capabilities

List solve modes and every public capability.

Parameters

No parameters.

Raw schema

{
  "type": "object",
  "properties": {},
  "title": "get_capabilitiesArguments"
}

get_settings

Read this caller's saved solve and safety settings.

Parameters1

user_id

any

optional

Raw schema

{
  "type": "object",
  "properties": {
    "user_id": {
      "anyOf": [
        {
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "User Id"
    }
  },
  "title": "get_settingsArguments"
}

set_settings

Save caller-scoped settings from one dict (depth, optimization, trace, relay, governance); keys and values in get_capabilities.

Parameters2

settings	any	optional
user_id	any	optional

Raw schema

{
  "type": "object",
  "properties": {
    "settings": {
      "anyOf": [
        {
          "additionalProperties": true,
          "type": "object"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Settings"
    },
    "user_id": {
      "anyOf": [
        {
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "User Id"
    }
  },
  "title": "set_settingsArguments"
}

get_result_detail

Fetch the full caller-owned result behind a lean solve.

Parameters1

result_id

string

required

Raw schema

{
  "type": "object",
  "properties": {
    "result_id": {
      "title": "Result Id",
      "type": "string"
    }
  },
  "required": [
    "result_id"
  ],
  "title": "get_result_detailArguments"
}

export_result

Export a caller-owned result in a supported format.

Parameters2

result_id	string	required
format	string	optional

Raw schema

{
  "type": "object",
  "properties": {
    "result_id": {
      "title": "Result Id",
      "type": "string"
    },
    "format": {
      "default": "json",
      "title": "Format",
      "type": "string"
    }
  },
  "required": [
    "result_id"
  ],
  "title": "export_resultArguments"
}

compress_context

Compress long context into a task-focused brief; use before reading any source over 8,000 characters.

Parameters3

text	string	required
task	string	required
retention	string	optional

Raw schema

{
  "type": "object",
  "properties": {
    "text": {
      "title": "Text",
      "type": "string"
    },
    "task": {
      "title": "Task",
      "type": "string"
    },
    "retention": {
      "default": "task_local",
      "title": "Retention",
      "type": "string"
    }
  },
  "required": [
    "text",
    "task"
  ],
  "title": "compress_contextArguments"
}

delegation

Create, resume, or report a bounded subagent packet (action='create'|'resume'|'report').

Parameters13

action	string	required
delegation_id	string	optional
handoff_id	string	optional
task	string	optional
role	string	optional
owned_paths	any	optional
read_only_inputs	any	optional
acceptance_criteria	any	optional
stop_conditions	any	optional
outcome	string	optional
changed_paths	any	optional
test_evidence	any	optional
unresolved	any	optional

Raw schema

{
  "type": "object",
  "properties": {
    "action": {
      "title": "Action",
      "type": "string"
    },
    "delegation_id": {
      "default": "",
      "title": "Delegation Id",
      "type": "string"
    },
    "handoff_id": {
      "default": "",
      "title": "Handoff Id",
      "type": "string"
    },
    "task": {
      "default": "",
      "title": "Task",
      "type": "string"
    },
    "role": {
      "default": "",
      "title": "Role",
      "type": "string"
    },
    "owned_paths": {
      "anyOf": [
        {
          "items": {
            "type": "string"
          },
          "type": "array"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Owned Paths"
    },
    "read_only_inputs": {
      "anyOf": [
        {
          "items": {
            "type": "string"
          },
          "type": "array"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Read Only Inputs"
    },
    "acceptance_criteria": {
      "anyOf": [
        {
          "items": {
            "type": "string"
          },
          "type": "array"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Acceptance Criteria"
    },
    "stop_conditions": {
      "anyOf": [
        {
          "items": {
            "type": "string"
          },
          "type": "array"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Stop Conditions"
    },
    "outcome": {
      "default": "",
      "title": "Outcome",
      "type": "string"
    },
    "changed_paths": {
      "anyOf": [
        {
          "items": {
            "type": "string"
          },
          "type": "array"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Changed Paths"
    },
    "test_evidence": {
      "anyOf": [
        {
          "items": {
            "type": "string"
          },
          "type": "array"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Test Evidence"
    },
    "unresolved": {
      "anyOf": [
        {
          "items": {
            "type": "string"
          },
          "type": "array"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Unresolved"
    }
  },
  "required": [
    "action"
  ],
  "title": "delegationArguments"
}

handoff

Create, safely resume, or list caller-scoped cross-session checkpoints (action='create'|'resume'|'list'). Create accepts an optional subject and source_sessions; Liminality assigns the subject number. Resume by handoff_id — 'latest' resumes the newest unfinished resumable checkpoint, or asks which one when several were created in the last 24h; detail='full' returns the complete checkpoint.

Parameters14

action	string	required
task	string	optional
session_state	string	optional
evidence	string	optional
checkpoint	any	optional
handoff_id	string	optional
parent_handoff_id	string	optional
current_artifact_hash	string	optional
resources	string	optional
tool_manifest	any	optional
detail	string	optional
subject	string	optional
source_sessions	any	optional
refs	any	optional

Raw schema

{
  "type": "object",
  "properties": {
    "action": {
      "title": "Action",
      "type": "string"
    },
    "task": {
      "default": "",
      "title": "Task",
      "type": "string"
    },
    "session_state": {
      "default": "",
      "title": "Session State",
      "type": "string"
    },
    "evidence": {
      "default": "",
      "title": "Evidence",
      "type": "string"
    },
    "checkpoint": {
      "anyOf": [
        {
          "additionalProperties": true,
          "type": "object"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Checkpoint"
    },
    "handoff_id": {
      "default": "",
      "title": "Handoff Id",
      "type": "string"
    },
    "parent_handoff_id": {
      "default": "",
      "title": "Parent Handoff Id",
      "type": "string"
    },
    "current_artifact_hash": {
      "default": "",
      "title": "Current Artifact Hash",
      "type": "string"
    },
    "resources": {
      "default": "",
      "title": "Resources",
      "type": "string"
    },
    "tool_manifest": {
      "anyOf": [
        {
          "items": {
            "additionalProperties": true,
            "type": "object"
          },
          "type": "array"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Tool Manifest"
    },
    "detail": {
      "default": "lean",
      "title": "Detail",
      "type": "string"
    },
    "subject": {
      "default": "",
      "title": "Subject",
      "type": "string"
    },
    "source_sessions": {
      "anyOf": [
        {
          "items": {
            "type": "string"
          },
          "type": "array"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Source Sessions"
    },
    "refs": {
      "anyOf": [
        {
          "additionalProperties": true,
          "type": "object"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Refs"
    }
  },
  "required": [
    "action"
  ],
  "title": "handoffArguments"
}

ask_form

Create a short form for missing user-specific facts.

Parameters3

question	string	required
max_depth	integer	optional
verify_divergence	boolean	optional

Raw schema

{
  "type": "object",
  "properties": {
    "question": {
      "title": "Question",
      "type": "string"
    },
    "max_depth": {
      "default": 3,
      "title": "Max Depth",
      "type": "integer"
    },
    "verify_divergence": {
      "default": false,
      "title": "Verify Divergence",
      "type": "boolean"
    }
  },
  "required": [
    "question"
  ],
  "title": "ask_formArguments"
}

apply_form

Apply form answers to sharpen a task.

Parameters2

question	string	required
fills	object	required

Raw schema

{
  "type": "object",
  "properties": {
    "question": {
      "title": "Question",
      "type": "string"
    },
    "fills": {
      "additionalProperties": true,
      "title": "Fills",
      "type": "object"
    }
  },
  "required": [
    "question",
    "fills"
  ],
  "title": "apply_formArguments"
}

report_outcome

Report verified real-world outcome evidence for a result.

Parameters13

result_id	string	optional
worked	boolean	optional
note	string	optional
intent	string	optional
leaf	string	optional
endpoint	string	optional
endpoint_correct	boolean	optional
read_back	string	optional
efficiency_note	string	optional
could_have_done_better	string	optional
human_verdict	string	optional
completion_event	string	optional
evidence_ref	string	optional

Raw schema

{
  "type": "object",
  "properties": {
    "result_id": {
      "type": "string",
      "minLength": 1
    },
    "worked": {
      "type": "boolean"
    },
    "note": {
      "type": "string",
      "minLength": 1
    },
    "intent": {
      "type": "string",
      "minLength": 1
    },
    "leaf": {
      "type": "string",
      "minLength": 1
    },
    "endpoint": {
      "type": "string",
      "minLength": 1
    },
    "endpoint_correct": {
      "type": "boolean"
    },
    "read_back": {
      "type": "string",
      "minLength": 1
    },
    "efficiency_note": {
      "type": "string"
    },
    "could_have_done_better": {
      "type": "string"
    },
    "human_verdict": {
      "type": "string"
    },
    "completion_event": {
      "type": "string"
    },
    "evidence_ref": {
      "type": "string"
    }
  },
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://physea.ai/schemas/liminality-report-outcome-v1.schema.json",
  "anyOf": [
    {
      "required": [
        "result_id",
        "worked"
      ],
      "anyOf": [
        {
          "required": [
            "note"
          ]
        },
        {
          "required": [
            "read_back"
          ]
        }
      ]
    },
    {
      "required": [
        "intent",
        "leaf",
        "endpoint",
        "endpoint_correct"
      ]
    },
    {
      "required": [
        "intent",
        "leaf",
        "endpoint",
        "completion_event",
        "evidence_ref"
      ]
    },
    {
      "required": [
        "intent",
        "leaf",
        "endpoint",
        "human_verdict",
        "read_back"
      ]
    }
  ],
  "additionalProperties": false
}

report_feedback

Report structured feedback on a result.

Parameters7

rating	any	optional
helpful	any	optional
comment	string	optional
missing	string	optional
tool	string	optional
route_key	string	optional
user_id	any	optional

Raw schema

{
  "type": "object",
  "properties": {
    "rating": {
      "anyOf": [
        {
          "type": "integer"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Rating"
    },
    "helpful": {
      "anyOf": [
        {
          "type": "boolean"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Helpful"
    },
    "comment": {
      "default": "",
      "title": "Comment",
      "type": "string"
    },
    "missing": {
      "default": "",
      "title": "Missing",
      "type": "string"
    },
    "tool": {
      "default": "",
      "title": "Tool",
      "type": "string"
    },
    "route_key": {
      "default": "",
      "title": "Route Key",
      "type": "string"
    },
    "user_id": {
      "anyOf": [
        {
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "User Id"
    }
  },
  "title": "report_feedbackArguments"
}

apply_grade

Apply a grade-form answer to a result.

Parameters3

result_id	string	required
choice	string	required
note	string	optional

Raw schema

{
  "type": "object",
  "properties": {
    "result_id": {
      "title": "Result Id",
      "type": "string"
    },
    "choice": {
      "title": "Choice",
      "type": "string"
    },
    "note": {
      "default": "",
      "title": "Note",
      "type": "string"
    }
  },
  "required": [
    "result_id",
    "choice"
  ],
  "title": "apply_gradeArguments"
}

optimize

Rank bounded one-factor variants for a route against your stated objective; candidates remain unsaved.

Parameters4

target	string	required
objective	string	required
budget	integer	optional
top_k	integer	optional

Raw schema

{
  "type": "object",
  "properties": {
    "target": {
      "title": "Target",
      "type": "string"
    },
    "objective": {
      "title": "Objective",
      "type": "string"
    },
    "budget": {
      "default": 5,
      "title": "Budget",
      "type": "integer"
    },
    "top_k": {
      "default": 5,
      "title": "Top K",
      "type": "integer"
    }
  },
  "required": [
    "target",
    "objective"
  ],
  "title": "optimizeArguments"
}

algorithms

Create, validate, release and run pure algorithms over versioned datasets; runs replay byte-identical with zero model calls. action='run' is one solve; other actions are free.

Parameters14

action	string	required
spec	any	optional
version_id	string	optional
changes	any	optional
dataset_id	string	optional
name	string	optional
columns	any	optional
rows	any	optional
key	string	optional
parameters	any	optional
label	string	optional
run_id	string	optional
run_ids	any	optional
correction	any	optional

Raw schema

{
  "type": "object",
  "properties": {
    "action": {
      "title": "Action",
      "type": "string"
    },
    "spec": {
      "anyOf": [
        {
          "additionalProperties": true,
          "type": "object"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Spec"
    },
    "version_id": {
      "default": "",
      "title": "Version Id",
      "type": "string"
    },
    "changes": {
      "anyOf": [
        {
          "additionalProperties": true,
          "type": "object"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Changes"
    },
    "dataset_id": {
      "default": "",
      "title": "Dataset Id",
      "type": "string"
    },
    "name": {
      "default": "",
      "title": "Name",
      "type": "string"
    },
    "columns": {
      "anyOf": [
        {
          "items": {
            "additionalProperties": true,
            "type": "object"
          },
          "type": "array"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Columns"
    },
    "rows": {
      "anyOf": [
        {
          "items": {
            "additionalProperties": true,
            "type": "object"
          },
          "type": "array"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Rows"
    },
    "key": {
      "default": "",
      "title": "Key",
      "type": "string"
    },
    "parameters": {
      "anyOf": [
        {
          "additionalProperties": true,
          "type": "object"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Parameters"
    },
    "label": {
      "default": "",
      "title": "Label",
      "type": "string"
    },
    "run_id": {
      "default": "",
      "title": "Run Id",
      "type": "string"
    },
    "run_ids": {
      "anyOf": [
        {
          "items": {
            "type": "string"
          },
          "type": "array"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Run Ids"
    },
    "correction": {
      "anyOf": [
        {
          "additionalProperties": true,
          "type": "object"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Correction"
    }
  },
  "required": [
    "action"
  ],
  "title": "algorithmsArguments"
}

task

Parameters8

action	string	optional
task_id	string	optional
changes	any	optional
version	integer	optional
fields	any	optional
limit	integer	optional
user_id	any	optional
expected_version	any	optional

Raw schema

{
  "type": "object",
  "properties": {
    "action": {
      "default": "get",
      "title": "Action",
      "type": "string"
    },
    "task_id": {
      "default": "",
      "title": "Task Id",
      "type": "string"
    },
    "changes": {
      "anyOf": [
        {
          "additionalProperties": true,
          "type": "object"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Changes"
    },
    "version": {
      "default": 0,
      "title": "Version",
      "type": "integer"
    },
    "fields": {
      "anyOf": [
        {
          "additionalProperties": true,
          "type": "object"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Fields"
    },
    "limit": {
      "default": 50,
      "title": "Limit",
      "type": "integer"
    },
    "user_id": {
      "anyOf": [
        {
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "User Id"
    },
    "expected_version": {
      "anyOf": [
        {
          "type": "integer"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Expected Version"
    }
  },
  "title": "taskArguments"
}

recover

Parameters13

action	string	optional
case_id	string	optional
fields	any	optional
candidate_id	string	optional
user_change	any	optional
outcome	string	optional
note	string	optional
repaired_claims	any	optional
criteria_ids	any	optional
candidate	any	optional
task_id	string	optional
limit	integer	optional
user_id	any	optional

Raw schema

{
  "type": "object",
  "properties": {
    "action": {
      "default": "list",
      "title": "Action",
      "type": "string"
    },
    "case_id": {
      "default": "",
      "title": "Case Id",
      "type": "string"
    },
    "fields": {
      "anyOf": [
        {
          "additionalProperties": true,
          "type": "object"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Fields"
    },
    "candidate_id": {
      "default": "",
      "title": "Candidate Id",
      "type": "string"
    },
    "user_change": {
      "anyOf": [
        {
          "additionalProperties": true,
          "type": "object"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "User Change"
    },
    "outcome": {
      "default": "",
      "title": "Outcome",
      "type": "string"
    },
    "note": {
      "default": "",
      "title": "Note",
      "type": "string"
    },
    "repaired_claims": {
      "anyOf": [
        {
          "items": {
            "additionalProperties": true,
            "type": "object"
          },
          "type": "array"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Repaired Claims"
    },
    "criteria_ids": {
      "anyOf": [
        {
          "items": {
            "type": "string"
          },
          "type": "array"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Criteria Ids"
    },
    "candidate": {
      "anyOf": [
        {
          "additionalProperties": true,
          "type": "object"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Candidate"
    },
    "task_id": {
      "default": "",
      "title": "Task Id",
      "type": "string"
    },
    "limit": {
      "default": 50,
      "title": "Limit",
      "type": "integer"
    },
    "user_id": {
      "anyOf": [
        {
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "User Id"
    }
  },
  "title": "recoverArguments"
}

runs

Parameters17

action	string	optional
run_id	string	optional
operation	any	optional
idempotency_key	string	optional
mode	string	optional
task_id	string	optional
node_id	string	optional
policy	any	optional
attempt_id	string	optional
outcome	any	optional
limit	integer	optional
user_id	any	optional
approval_id	string	optional
operation_kinds	any	optional
danger_classes	any	optional
hours	number	optional
reason	string	optional

Raw schema

{
  "type": "object",
  "properties": {
    "action": {
      "default": "status",
      "title": "Action",
      "type": "string"
    },
    "run_id": {
      "default": "",
      "title": "Run Id",
      "type": "string"
    },
    "operation": {
      "anyOf": [
        {
          "additionalProperties": true,
          "type": "object"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Operation"
    },
    "idempotency_key": {
      "default": "",
      "title": "Idempotency Key",
      "type": "string"
    },
    "mode": {
      "default": "live",
      "title": "Mode",
      "type": "string"
    },
    "task_id": {
      "default": "",
      "title": "Task Id",
      "type": "string"
    },
    "node_id": {
      "default": "",
      "title": "Node Id",
      "type": "string"
    },
    "policy": {
      "anyOf": [
        {
          "additionalProperties": true,
          "type": "object"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Policy"
    },
    "attempt_id": {
      "default": "",
      "title": "Attempt Id",
      "type": "string"
    },
    "outcome": {
      "anyOf": [
        {
          "additionalProperties": true,
          "type": "object"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Outcome"
    },
    "limit": {
      "default": 50,
      "title": "Limit",
      "type": "integer"
    },
    "user_id": {
      "anyOf": [
        {
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "User Id"
    },
    "approval_id": {
      "default": "",
      "title": "Approval Id",
      "type": "string"
    },
    "operation_kinds": {
      "anyOf": [
        {
          "items": {
            "type": "string"
          },
          "type": "array"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Operation Kinds"
    },
    "danger_classes": {
      "anyOf": [
        {
          "items": {
            "type": "string"
          },
          "type": "array"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Danger Classes"
    },
    "hours": {
      "default": 24,
      "title": "Hours",
      "type": "number"
    },
    "reason": {
      "default": "",
      "title": "Reason",
      "type": "string"
    }
  },
  "title": "runsArguments"
}

research

Run a multi-source research pass for a question.

Parameters2

question	string	required
max_depth	integer	optional

Raw schema

{
  "type": "object",
  "properties": {
    "question": {
      "title": "Question",
      "type": "string"
    },
    "max_depth": {
      "default": 3,
      "title": "Max Depth",
      "type": "integer"
    }
  },
  "required": [
    "question"
  ],
  "title": "researchArguments"
}

composio_connect

Create a caller-scoped connection link for a toolkit.

Parameters2

toolkit	string	required
user_id	any	optional

Raw schema

{
  "type": "object",
  "properties": {
    "toolkit": {
      "title": "Toolkit",
      "type": "string"
    },
    "user_id": {
      "anyOf": [
        {
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "User Id"
    }
  },
  "required": [
    "toolkit"
  ],
  "title": "composio_connectArguments"
}

register_asset

Parameters4

name	string	required
kind	string	optional
attributes	any	optional
user_id	any	optional

Raw schema

{
  "type": "object",
  "properties": {
    "name": {
      "title": "Name",
      "type": "string"
    },
    "kind": {
      "default": "",
      "title": "Kind",
      "type": "string"
    },
    "attributes": {
      "anyOf": [
        {
          "additionalProperties": true,
          "type": "object"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Attributes"
    },
    "user_id": {
      "anyOf": [
        {
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "User Id"
    }
  },
  "required": [
    "name"
  ],
  "title": "register_assetArguments"
}

set_priorities

Save caller-scoped tasks and importance signals.

Parameters2

items	any	required
user_id	any	optional

Raw schema

{
  "type": "object",
  "properties": {
    "items": {
      "anyOf": [
        {
          "items": {
            "additionalProperties": true,
            "type": "object"
          },
          "type": "array"
        },
        {
          "additionalProperties": true,
          "type": "object"
        }
      ],
      "title": "Items"
    },
    "user_id": {
      "anyOf": [
        {
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "User Id"
    }
  },
  "required": [
    "items"
  ],
  "title": "set_prioritiesArguments"
}

taxis

Return this caller's deterministic ranked priority frame.

Parameters1

user_id

any

optional

Raw schema

{
  "type": "object",
  "properties": {
    "user_id": {
      "anyOf": [
        {
          "type": "string"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "User Id"
    }
  },
  "title": "taxisArguments"
}

catalog

Read the capability catalog: action='search' returns every saved candidate for an objective with distinctiveness deltas; action='delta' compares a named entity pair.

Parameters4

action	string	optional
objective	string	optional
entity_a	string	optional
entity_b	string	optional

Raw schema

{
  "type": "object",
  "properties": {
    "action": {
      "default": "search",
      "title": "Action",
      "type": "string"
    },
    "objective": {
      "default": "",
      "title": "Objective",
      "type": "string"
    },
    "entity_a": {
      "default": "",
      "title": "Entity A",
      "type": "string"
    },
    "entity_b": {
      "default": "",
      "title": "Entity B",
      "type": "string"
    }
  },
  "title": "catalogArguments"
}

setup

Context-minimality setup with no questions (auto); or profiles, status, verify.

Parameters4

action	string	optional
profile	string	optional
use_case	string	optional
first_request	string	optional

Raw schema

{
  "type": "object",
  "properties": {
    "action": {
      "default": "auto",
      "title": "Action",
      "type": "string"
    },
    "profile": {
      "default": "",
      "title": "Profile",
      "type": "string"
    },
    "use_case": {
      "default": "",
      "title": "Use Case",
      "type": "string"
    },
    "first_request": {
      "default": "",
      "title": "First Request",
      "type": "string"
    }
  },
  "title": "setupArguments"
}

manage

Compatibility router for retained-context lifecycle actions.

Parameters2

action	string	required
payload	any	optional

Raw schema

{
  "type": "object",
  "properties": {
    "action": {
      "title": "Action",
      "type": "string"
    },
    "payload": {
      "anyOf": [
        {
          "additionalProperties": true,
          "type": "object"
        },
        {
          "type": "null"
        }
      ],
      "default": null,
      "title": "Payload"
    }
  },
  "required": [
    "action"
  ],
  "title": "manageArguments"
}

README.md5 KBView on GitHub

Liminal

Augment your work with Liminal. Stop repeating yourself and build on what’s already solved.

Docs · Get a free key · Examples

smithery badge MCP Badge

Liminal turns objectives into structured work by strategizing through an atomic, deterministic basis: setting guardrails around scope, compiling resources and constraints, uncovering gaps, and organizing research and next steps. All while carrying continuity and priorities forward so your AI builds on what’s already done.

Connect easily through MCP or use the desktop app.

Install

Remote server, nothing to build or run locally. Get a key at https://physea.ai/signup, then add it to your client. Ready-to-copy files for each client are in examples/, with a first solve walkthrough.

Claude Code

bash

claude mcp add liminal --transport http https://liminality.physea.ai/mcp --header "X-API-Key: YOUR_KEY"

Cursor

Add to ~/.cursor/mcp.json:

json

{
  "mcpServers": {
    "liminal": {
      "url": "https://liminality.physea.ai/mcp",
      "headers": { "X-API-Key": "YOUR_KEY" }
    }
  }
}

VS Code

Add to .vscode/mcp.json:

json

{
  "servers": {
    "liminal": {
      "type": "http",
      "url": "https://liminality.physea.ai/mcp",
      "headers": { "X-API-Key": "YOUR_KEY" }
    }
  }
}

Endpoint: https://liminality.physea.ai/mcp
Transport: streamable HTTP (remote, hosted)
Auth: API key via X-API-Key or Authorization: Bearer, or OAuth 2.1.
Registry name: ai.physea/liminality (Official MCP Registry)

What it is

Liminal is a reasoning layer that sits between your agent and the tools it could call. Instead of answering from memory, it works out the structure of the request first, then routes each piece to something real.

It runs on context-dependent determinism. Same question with the same known context routes the same way every time, so a result is reproducible and you can hand the route to someone else and get the same thing back. Change the context, and the route adapts to it. The determinism is in how it decides, not a frozen cache of canned answers.

Tools

solve is the front door. Give it anything non-trivial. It works out the structure (decompose, ground to real tools, score the decision) and returns a worked result: a scored decision frame for a choice, or a grounded answer for a question.
research runs a deeper multi-source pass that pulls real information for a question.
ask_form / apply_form: when the answer depends on your specifics, it hands back a short multiple-choice form. Relay it, then apply_form folds the answers into a sharper result.
get_my_context: what's known about you and what you've connected.
register_asset / set_preference: tell it about your material and how you like results.
report_feedback / report_outcome: tell it how a result did, so routes improve with use.
composio_connect: connect a tool so an action can run against it.

Discovery

https://liminality.physea.ai/.well-known/mcp
https://liminality.physea.ai/.well-known/agent.json
https://liminality.physea.ai/llms.txt

Links

Site and docs: https://physea.ai/mcp
Get a key: https://physea.ai/signup
Privacy: https://physea.ai/privacy-policy
By Physea

© 2026 Physea Labs. All rights reserved. "Liminal", the name, and the logo are trademarks of Physea. This repository is a listing for a hosted service; no rights to the service, its software, or its data are granted. Use of the service is governed by the terms at https://physea.ai/mcp.

Install

Tools (27)

solveget_capabilitiesget_settingsset_settingsget_result_detailexport_resultcompress_contextdelegation

Remote endpoint

Streamable HTTP

Hosted server - connect over the network, no local install.

url

https://liminality.physea.ai/mcp

Add to VS Code Add to Cursor

claude_desktop_config.json

json

{
  "mcpServers": {
    "liminality": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://liminality.physea.ai/mcp"
      ]
    }
  }
}

Desktop config is stdio-only; this bridges via mcp-remote. Native remote: Settings > Connectors.

ai.physea/liminality MCP Server

🛠️ Key Features

🚀 Use Cases

⚡ Developer Benefits

⚠️ Limitations

Topics

solve

get_capabilities

get_settings

set_settings

get_result_detail

export_result

compress_context

delegation

handoff

ask_form

apply_form

report_outcome

report_feedback

apply_grade

optimize

algorithms

task

recover

runs

research

composio_connect

register_asset

set_priorities

taxis

catalog

setup

manage

Liminal

Install

What it is

Tools

Discovery

Links

Install

Remote endpoint

Related servers