delimit

Official

Overview Schema Source Related Servers Discussions Admin

Delimit Session Handoff

delimit_session_handoff

Preserve session context by writing a structured handoff record with decisions, blockers, and next actions, so a future session can resume where the last one left off.

Instructions

Save a session summary for cross-session continuity.

When to use: at the end of a productive session, to leave a structured record the next session can recover. When NOT to use: for richer cross-model state (use delimit_soul_capture, which auto-detects more) or single-line memory (delimit_memory_store).

Sibling contrast: delimit_soul_capture writes a state soul directly; this writes a structured handoff with explicit fields. Session Phoenix, when installed, enriches soul capture with git and transcript state.

Side effects: performs the ONE canonical session close via ai.ledger_manager.session_handoff: refreshes the soul, writes the handoff record, creates the handoff receipt, then reads every written record back and reports verified (LED-5321 M3). Identifier/path fields retain legacy comma-list coercion. Prose decisions and blockers preserve commas in a plain string; native/JSON lists carry multiple entries. LED-3731: also refreshes a lightweight pointer-soul for project_path (default = cwd) so the NEXT delimit_revive for that project returns THIS handoff's state rather than a stale older soul.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`summary`	Yes	2-3 sentence summary of the session. Required.
`venture`	No	Venture context. Empty = auto-detect.
`blockers`	No	What's blocked and why.
`priority`	No	Receipt priority (P0/P1/P2). Default P1.	P1
`to_model`	No	Receipt recipient model. Default any.	any
`completed`	No	Receipt completed items. Empty = reuse items_completed.
`assumptions`	No	Receipt assumptions.
`items_added`	No	Newly added item ids as list or comma string.
`next_action`	No	Receipt next action for the receiving session.
`project_path`	No	Project path whose revive soul this handoff should refresh. Empty = auto-detect from cwd.
`source_model`	No	Closing model identity. Empty = auto-detect. Any model may close; no supervisor is assumed.
`files_changed`	No	Key files that were modified.
`key_decisions`	No	Key decisions or consensus results.
`not_completed`	No	Receipt not-completed items. Empty = reuse items_added.
`items_completed`	No	Completed ledger item ids (e.g. ["LED-164"]) as list or comma string.
`task_description`	No	Receipt task line. Empty = reuse summary.
`logical_session_id`	No	Idempotency key for the canonical close. Re-closing with the same id updates/returns the same soul+handoff+receipt instead of duplicating. Empty = fresh close.

Output Schema

TableJSON Schema

Name	Required	Description	Default
No arguments

Schema Changelog

Changes observed during successful MCP inspections.

Changed9 schema fields changedv4.19.1Sep 15, 2026

addedInput schema / properties / assumptions

Added value: +{
+  "anyOf": [
+    {
+      "type": "string"
+    },
+    {
+      "items": {
+        "type": "string"
+      },
+      "type": "array"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Receipt assumptions."
+}

addedInput schema / properties / completed

Added value: +{
+  "anyOf": [
+    {
+      "type": "string"
+    },
+    {
+      "items": {
+        "type": "string"
+      },
+      "type": "array"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Receipt completed items. Empty = reuse items_completed."
+}

addedInput schema / properties / logical_session_id

Added value: +{
+  "default": "",
+  "description": "Idempotency key for the canonical close. Re-closing with the same id updates/returns the same soul+handoff+receipt instead of duplicating. Empty = fresh close.",
+  "type": "string"
+}

addedInput schema / properties / next_action

Added value: +{
+  "default": "",
+  "description": "Receipt next action for the receiving session.",
+  "type": "string"
+}

addedInput schema / properties / not_completed

Added value: +{
+  "anyOf": [
+    {
+      "type": "string"
+    },
+    {
+      "items": {
+        "type": "string"
+      },
+      "type": "array"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "Receipt not-completed items. Empty = reuse items_added."
+}

addedInput schema / properties / priority

Added value: +{
+  "default": "P1",
+  "description": "Receipt priority (P0/P1/P2). Default P1.",
+  "type": "string"
+}

addedInput schema / properties / source_model

Added value: +{
+  "default": "",
+  "description": "Closing model identity. Empty = auto-detect. Any model may close; no supervisor is assumed.",
+  "type": "string"
+}

addedInput schema / properties / task_description

Added value: +{
+  "default": "",
+  "description": "Receipt task line. Empty = reuse summary.",
+  "type": "string"
+}

addedInput schema / properties / to_model

Added value: +{
+  "default": "any",
+  "description": "Receipt recipient model. Default any.",
+  "type": "string"
+}

Changed1 schema field changedv4.13.2Sep 5, 2026

addedInput schema / properties / project_path

Added value: +{
+  "default": "",
+  "description": "Project path whose revive soul this handoff should refresh. Empty = auto-detect from cwd.",
+  "type": "string"
+}

Changed4 schema fields changedv4.7.9Jun 9, 2026

changedInput schema / properties / items_added / description

Previous value: -"List of newly added item IDs."New value: +"Newly added item ids as list or comma string."

changedInput schema / properties / items_completed / description

Previous value: -"List of completed ledger item IDs (e.g. [\"LED-164\", \"LED-165\"])."New value: +"Completed ledger item ids (e.g. [\"LED-164\"]) as list or comma string."

changedInput schema / properties / summary / description

Previous value: -"2-3 sentence summary of what happened this session."New value: +"2-3 sentence summary of the session. Required."

changedInput schema / properties / venture / description

Previous value: -"Venture context. Auto-detects if empty."New value: +"Venture context. Empty = auto-detect."

Addedv4.5.5May 7, 2026

TDQS

A4.7/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

The description discloses extensive behavioral detail beyond the annotations (readOnlyHint=false, destructiveHint=false). It explains the canonical session close, the sequence of actions (refreshes soul, writes handoff, creates receipt, reads back, reports verified), references ledger IDs (LED-5321 M3, LED-3731), and describes legacy comma-list coercion and pointer-soul refresh. This goes well beyond the minimal annotation information and gives the agent a clear picture of side effects.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is well-organized with clear sections ('When to use', 'When NOT to use', 'Sibling contrast', 'Side effects') and avoids redundancy. While it is somewhat long, every sentence adds substantive value, and the structure aids readability. It could be slightly more compact but is far from bloated.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's complexity (17 parameters, 1 required) and the presence of an output schema, the description is comprehensive. It covers usage context, side effects, and parameter nuances. The description doesn't explicitly mention error cases or return format, but the output schema covers return values, so this is acceptable. It is complete enough for an agent to invoke the tool correctly.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the schema documents all parameters. The description adds extra meaning for specific parameters: it explains that identifier/path fields retain comma-list coercion while prose fields preserve commas, and that logical_session_id provides idempotency. This adds context beyond what the schema provides, warranting a score above the baseline of 3.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's purpose: 'Save a session summary for cross-session continuity.' It uses a specific verb and resource, and differentiates itself from siblings delimit_soul_capture and delimit_memory_store with explicit contrast. An agent can immediately understand what this tool does and how it differs from similar ones.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description explicitly states when to use ('at the end of a productive session') and when NOT to use (for richer cross-model state use delimit_soul_capture, for single-line memory use delimit_memory_store). It also provides sibling contrast. This leaves no ambiguity about when to select this tool over alternatives.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server