Task List Batch Editor

productivity.todo_update

ProductionBounded custom (Class B)Deterministicstrictfastv1.0.0

No network

Overview

You send a task list and a list of edits, and we hand back the list as it stands after those edits plus a per-operation receipt saying what each one did. Targets are named exactly, by identity, by position or by whole title, and never by a pattern or a similarity score, so an edit either hits the item you meant or tells you it hit nothing. Nothing outside the response changes: there is no tracker behind this, and the value we return is the whole of the result.

An agent that edits a checklist by rewriting Markdown loses the identities that told it which item was which, and the next read cannot tell an edit from a coincidence.

Reliability

Same list, same operations, same bytes back. Replaying a batch converges: a completion already applied comes back as ALREADY_IN_STATE, and an addition already present is refused unless you asked for duplicates. Failures, none worth retrying unchanged: TODO_UPDATE_LIST_INVALID means an entry was missing its id, text or state, carried an unexpected property, or ran past a stated cap, so fix that entry; TODO_UPDATE_ITEM_ID_DUPLICATE means two entries claimed one identity, and the response names the position of the second; TODO_UPDATE_OPERATIONS_INVALID means an operation was malformed, most often an ADD or RETITLE with no text, so correct the operation named by index; TODO_UPDATE_OPERATION_UNSUPPORTED means op was outside COMPLETE, REOPEN, ADD, REMOVE and RETITLE, so pick one of those five; TODO_UPDATE_MATCH_MODE_UNSUPPORTED means match.by was outside ID, ORDINAL and TITLE_EXACT, and there is deliberately no fuzzy mode to fall back to; TODO_UPDATE_MATCH_VALUE_INVALID means the value did not suit the mode, such as a title where a position belongs, so send the right shape; TODO_UPDATE_TARGET_NOT_FOUND means an operation matched nothing under the default policy, so refresh the list or set missingTargetPolicy to RECORD; TODO_UPDATE_TARGET_AMBIGUOUS means a title matched more than one item, and the count comes back so you can switch that operation to match by identity; TODO_UPDATE_DUPLICATE_TEXT means an addition repeats text already in the list, so drop it or set allowDuplicateText to true; TODO_UPDATE_LIMIT_EXCEEDED means the list or the batch went past 20,000 items or 5,000 operations, so split the work; TODO_UPDATE_WORK_REFUSED means list size and batch size together project past our work budget, and both numbers come back so you know how far to trim; TODO_UPDATE_OUTPUT_TOO_LARGE means the resulting list would not fit in one inline response, so split the list and edit each part.

Context footprint

Returns the whole edited list plus one receipt line per operation, so a 500-item list comes back at 500 items. Item text is echoed because a list without its wording cannot be stored back. A successful call costs 2 execution credits.

When to use it

When you hold a task list as data and want a set of edits applied to it under exact matching, with a receipt saying what each edit actually did.

When you want the change written somewhere, or when you can only describe the target approximately.

Scope

Note

Apply an ordered batch of exact-match edits to a task list value and return the resulting list together with a per-operation receipt.

Non-goals

Does not connect to any tracker, calendar or note store
Does not match targets by pattern, prefix or similarity score
Does not reorder, sort or reparent items beyond the edits requested
Does not merge two lists or resolve a conflict between them

Properties

IdempotentRetry safeStreaming supportedFile input supportedRequires external service

Capabilities

data-transformation

Cost & billing

Flat rate per invocation

MetercallsBillable onsuccessCost per call2 credits

File handling

Accepts file_refProduces file_refMax input2 MB

How we handle your data

Sensitive output possible

What this means

This skill may receive sensitive content and may return submitted, transformed, extracted, or reformatted content.

Squillify guarantees

Runs inside Squillify
This skill runs inside Squillify.
Payloads are not sent to an external service by this skill.
Squillify does not permanently store submitted payloads.
Telemetry records metadata only, not submitted content.
Submitted payloads are not used for training.
Submitted payloads are not reviewed by humans.
This skill does not target sensitive data.
The response may contain submitted or transformed content.

Your responsibility

Squillify does not determine whether submitted content is sensitive. Your application is responsible for deciding what data it is allowed to submit. Your application is responsible for securing the response.

Read the Microskill Data Handling Policy

Input schema

Field	Type	Required	Description
todoList	array	Yes	List to edit, 0 through 20000 closed items each with unique id, state and text, plus optional extract extras.
operations	array	Yes	Edits to apply in order, 0 through 5000 closed operations each with op and, except for ADD, a match.
documentId	string	No	Document prefix for derived identities, 1 to 64 letters, digits, dots, underscores, colons or hyphens; omit it so identities start with a bare colon.
missingTargetPolicy	string	No	FAIL abandons the batch when a match hits nothing; RECORD keeps going. Omit it for FAIL.
allowDuplicateText	boolean	No	Whether ADD may repeat wording already in the list; omit it for false.

Output schema

Field	Type	Required	Description
items	array	Yes	List after every operation, 0 through 20000 closed items renumbered from zero without source offsets.
changes	array	Yes	One closed receipt per operation in send order, 0 through 5000 lines with sequence, code and involved identities.
summary	object	Yes	Closed batch counts: input and output sizes, applied, no-change and absent operations, plus checked and unchecked totals.

API contract

POST /v1/execute

curl

curl -sS -X POST "$BASE/v1/execute" -H "Authorization: Bearer $KEY" -H 'Content-Type: application/json' -d '{"skillKey":"productivity.todo_update","input":{"todoList":[{"id":"release-notes:b7cde273a37783e2:0","state":"UNCHECKED","text":"Draft the changelog"}],"operations":[{"op":"COMPLETE","match":{"by":"ORDINAL","value":0}}]}}'

Integrations

TypeScript SDK

import { Squillify } from "@squillify/sdk";

const sq = new Squillify({ apiKey: process.env.SQUILLIFY_API_KEY });

const { output } = await sq.execute("productivity.todo_update", {"todoList":[{"id":"release-notes:b7cde273a37783e2:0","ordinal":0,"depth":0,"state":"UNCHECKED","text":"Draft the changelog"}],"operations":[{"match":{"by":"ID","value":"release-notes:b7cde273a37783e2:0"},"op":"COMPLETE"}]});

Errors

Code	HTTP	Retryable	Description
VALIDATION_ERROR	400	No	The request payload failed schema validation.
UNAUTHORIZED	401	No	Missing or invalid API key.
SKILL_NOT_ACCESSIBLE	403	No	The API key is not granted access to this skill.
SKILL_NOT_AVAILABLE_FOR_PLAN	403	No	This skill is not included in your current plan.
PLAN_ELIGIBILITY_UNAVAILABLE	503	Yes	Plan eligibility could not be decided. Retry after backoff.
SKILL_NOT_FOUND	404	No	The requested skill or version does not exist.
TENANT_RATE_LIMITED	429	Yes	Request rate limit exceeded. Retry after backoff.
QUOTA_EXCEEDED	429	Yes	Plan quota exhausted. Retryable once the quota period resets.
EXECUTION_TIMEOUT	504	Yes	Execution exceeded the allowed time budget.
TODO_UPDATE_LIST_INVALID	400	No	The request payload failed this skill's input validation.
TODO_UPDATE_ITEM_ID_DUPLICATE	400	No	The request payload failed this skill's input validation.
TODO_UPDATE_OPERATIONS_INVALID	400	No	The request payload failed this skill's input validation.
TODO_UPDATE_OPERATION_UNSUPPORTED	400	No	The skill does not support the requested variant.
TODO_UPDATE_MATCH_MODE_UNSUPPORTED	400	No	The skill does not support the requested variant.
TODO_UPDATE_MATCH_VALUE_INVALID	400	No	The request payload failed this skill's input validation.
TODO_UPDATE_TARGET_NOT_FOUND	412	No	The input is valid but cannot be processed as-is.
TODO_UPDATE_TARGET_AMBIGUOUS	412	No	The input is valid but cannot be processed as-is.
TODO_UPDATE_DUPLICATE_TEXT	412	No	The input is valid but cannot be processed as-is.
TODO_UPDATE_LIMIT_EXCEEDED	429	No	A size, quota, rate, or time limit was exceeded.
TODO_UPDATE_WORK_REFUSED	429	No	A size, quota, rate, or time limit was exceeded.
TODO_UPDATE_OUTPUT_TOO_LARGE	429	No	A size, quota, rate, or time limit was exceeded.

Other limits

Limit	Value
Recommended inline bytes	32 KiB

Limitations

Does not connect to any tracker, calendar or note store
Does not match targets by pattern, prefix or similarity score
Does not reorder, sort or reparent items beyond the edits requested
Does not merge two lists or resolve a conflict between them
Editing a Markdown document in place, which this never touches
Finding the item you meant by rough description rather than exact identity

How to call

POST /v1/execute (requires authentication)

Request bodyjson

{
  "skillKey": "productivity.todo_update",
  "input": {
    "todoList": [
      {
        "id": "release-notes:b7cde273a37783e2:0",
        "ordinal": 0,
        "depth": 0,
        "state": "UNCHECKED",
        "text": "Draft the changelog"
      }
    ],
    "operations": [
      {
        "match": {
          "by": "ID",
          "value": "release-notes:b7cde273a37783e2:0"
        },
        "op": "COMPLETE"
      }
    ]
  }
}

Examples

Check off one item named by its identity

Inputjson

{
  "todoList": [
    {
      "id": "release-notes:b7cde273a37783e2:0",
      "ordinal": 0,
      "depth": 0,
      "state": "UNCHECKED",
      "text": "Draft the changelog"
    }
  ],
  "operations": [
    {
      "match": {
        "by": "ID",
        "value": "release-notes:b7cde273a37783e2:0"
      },
      "op": "COMPLETE"
    }
  ]
}

Expected outputjson

{
  "changes": [
    {
      "sequence": 0,
      "operation": "COMPLETE",
      "code": "COMPLETED",
      "itemId": "release-notes:b7cde273a37783e2:0",
      "ordinalBefore": 0,
      "ordinalAfter": 0,
      "stateBefore": "UNCHECKED",
      "stateAfter": "CHECKED"
    }
  ],
  "summary": {
    "inputCount": 1,
    "outputCount": 1,
    "appliedCount": 1,
    "noChangeCount": 0,
    "absentCount": 0,
    "checkedCount": 1,
    "uncheckedCount": 0
  },
  "items": [
    {
      "id": "release-notes:b7cde273a37783e2:0",
      "ordinal": 0,
      "depth": 0,
      "state": "CHECKED",
      "text": "Draft the changelog"
    }
  ]
}

Tip

The identity survives the edit, so a caller can apply the same operation again and read a no-change receipt.

Replayed completion plus a new item