AI DevelopmentMethodology6 min readPublished September 10, 2026

AI Tool Results: Which Details Should an Agent Keep?

Select AI tool results without losing evidence. Use a field-level reference for identifiers, errors, summaries and artifacts that agents can retrieve later.

DA
Digital Applied Team
AI research and implementation
Editorial dateSeptember 10, 2026
ReviewedSeptember 12, 2026

An AI tool result should give the agent enough evidence to choose its next action without forcing it to reread every record the tool touched. Keep exact values needed for decisions and follow-up calls, summarize relevant bulk information, and retain a usable path to the source material when more detail may be needed.

This reference proposes a field-level policy for tool responses. It does not recommend deleting the underlying records or using a universal token limit. The right response depends on the next task: a short answer may be enough for a status check, while reconciliation or editing needs exact identifiers and source details.

Key takeaways
  1. 01
    Preserve action-critical fields.Identifiers, units, status and task revision may be necessary even in a concise result.
  2. 02
    Summaries need provenance.Keep the source, scope and a retrievable path to omitted evidence.
  3. 03
    Context is not storage.Removing content from the model input does not delete the authoritative record.

01Reference methodDesign for the next decision

Ask what the model must decide after the tool returns. If the next action is to choose a matching record, provide discriminating fields. If it must update the record, include the exact identifier required by the update tool. If it must explain a failure, preserve the error category and affected fields.

Anthropic’s tool-design guidance recommends meaningful context and concise or detailed response formats. It also acknowledges that technical identifiers remain necessary for some downstream calls. Read that as a task-dependent design choice, not a blanket instruction to strip IDs.

A hypothetical inventory lookup might return the requested item, available quantity, unit and record ID. Returning the entire warehouse export would add irrelevant material. Returning only “in stock” could omit the unit or variant that determines whether the user can actually use it.

02Reference methodChoose a policy for each result field

The table below is a proposed design reference. Keep exact means preserve the value needed for the next operation or decision. Summarize means produce a bounded representation with provenance. Fetch on demand means retain a usable locator. Exclude refers to the model context, not automatic erasure from storage.

The rows are conditional. A full document may be essential for a document-editing task, and a resource identifier may be unnecessary for a one-off explanation. Define the required downstream operation before applying the policy; do not turn the table into an inflexible global filter.

Field groups in the proposed reference tableKeep exact6 fieldsSummarize3 fieldsFetch on demand3 fieldsExclude from model context3 fields
Groups and field counts from the proposed table below; these are not measured performance results.
Default treatmentField classCondition and reason
Keep exactOperation identifierNeeded to inspect or reconcile the same action; pair with a human-readable label.
Keep exactResource identifierKeep when a later tool requires it; avoid asking the model to reconstruct opaque IDs.
Keep exactTerminal statusPreserve success, failure, partial or uncertain state rather than a generic summary.
Keep exactTask revisionIdentify which authorized request the result answers.
Keep exactValue and unitKeep decision-relevant quantities with currency, timezone or other units.
Keep exactError categoryRetain the machine-actionable error or validation issue needed for the next step.
SummarizeRelevant record fieldsReturn the fields that answer the current question, with source identity.
SummarizeRepeated log messagesGroup repetition while preserving representative lines and the complete-log reference.
SummarizeSearch findingsInclude relevant excerpts, source URLs and uncertainty; do not imply exhaustive coverage.
Fetch on demandLong documentKeep a locator and relevant passage; retrieve more when the decision needs it.
Fetch on demandFull query resultStore in an authorized artifact; include row count, filters and a retrieval path.
Fetch on demandBinary artifactProvide path, format and version; load the actual artifact for tasks needing its content.
Exclude from model contextCredentialsUse trusted execution storage; return a redacted status instead of a secret.
Exclude from model contextIrrelevant personal fieldsDo not include data unrelated to the authorized task.
Exclude from model contextDuplicate raw payloadAvoid a second copy unless the protocol or a downstream consumer requires it.
Digital Applied proposed reference; primary-source distinctions are cited in the text. Reviewed September 12, 2026.

03Reference methodMake summaries explicit about what they omit

A useful summary states the question answered, the scope of the retrieval and the result. Include filters, source version or retrieval time when they affect the conclusion. If a query returned only its first page, the summary must not imply that no other records exist.

For long logs, preserve representative errors and a reference to the complete artifact. For retrieved documents, keep the passages that support the answer and enough source identity to open the same version. If the source cannot be retrieved later, be honest that a pointer alone cannot support follow-up work.

Anthropic’s context-engineering article describes selecting relevant context and retrieving information when needed. Our context reliability playbook covers the broader strategy. This reference focuses on the output contract that makes that strategy workable.

04Reference methodKeep protocol structure and application meaning aligned

The MCP tools specification dated June 18, 2025 supports structured content, unstructured content and resource links. A structured result is server-produced data; it is distinct from asking a model to generate schema-constrained output. Validate the actual returned data against the contract your application expects.

Use one authoritative status field and avoid a conflicting prose summary. If the structured result says an update failed, a friendly text block must not say it succeeded. Preserve whether the failure is a tool-execution error, a rejected argument or a connection problem so the next step can be appropriate.

The pinned MCP specification also recommends a serialized text representation of structured content for backward compatibility. Do not remove a protocol-required representation solely to save context. Deduplicate in the consumer’s model-input construction where supported, while preserving the tool response needed by other clients.

05Reference methodSeparate evidence, instructions and secrets

A retrieved page or document may contain text that looks like an instruction. Treat that as source content, not permission to change the application’s task. Selection can reduce exposure to irrelevant text, but a shorter payload does not make untrusted content safe.

Keep credentials in the trusted execution layer. A successful login operation can return a status and account label without returning its token. If the task needs personal information, include only the fields necessary for that authorized task and retain the appropriate access checks when retrieving more.

For voice frontends, OpenAI’s delegation guidance recommends keeping long records and full tool output in the backend. The voice layer needs concise relevant facts and status. Our background-tool voice guide applies that distinction to speaking while work continues.

06Reference methodTest what the compact result lets the agent do

Use the blank tool-result worksheet to list fields, their treatment, the downstream operation and the evidence retained. Compare a compact result with the current response on tasks that need exact identification, error recovery and follow-up detail. Do not evaluate token reduction alone.

Include a missing field, a stale version, a partial query and two similar resources. Check whether the agent requests more information when needed rather than guessing. A useful concise result can lead to another fetch; the goal is sufficient evidence for the task, not the smallest possible conversation.

The session-state responsibility reference helps decide where omitted records live. For implementation support, our AI transformation service can help define the tool contract and a bounded acceptance set. No context-selection experiment was run for this article.

Methodology

Evidence and scope

As-of date
September 12, 2026. September 10 is the editorial allocation; current documentation was reviewed later.
Method
Primary documentation and research were reviewed for the cited distinctions. Tables, worksheets and pilot checks are Digital Applied proposed methods, not observed deployment results.
Limitations
No production API workflow, vendor benchmark or participant study was executed for this article. Documentation can change; verify the selected configuration before implementation.

07Next stepPut the decision into practice

Put it into practice

Design for the next decision

Choose one noisy tool response and identify the next decision it must support. Preserve the exact fields that make that decision possible, then make every omission visible and recoverable where the task requires it.

From AI output to accepted work

Make your next AI workflow reviewable.

Define the result, the evidence and the people responsible for acceptance.

Clear scopePractical evaluationAccountable delivery
Implementation

Build around the result you need

  • Choose a representative workflow
  • Define acceptance evidence
  • Review the delivered outcome
Questions and answers

Applying the guide

Only when an identifier is needed for follow-up operations, provenance or reconciliation. Pair opaque IDs with readable labels when helpful.
Related dispatches

Continue reading

AI Development

AI Usage Is Rising: Is Your Team Completing More Work?

Assess rising AI usage against accepted work, review effort and delays. Build an evidence record before expanding access or claiming team productivity gains.

September 10, 2026 · 6 minRead
AI Development

Use Coding Agents to Build an Interactive Product Demo

Build an interactive product demo with coding agents. Define one user journey, label simulated behavior and test a resettable experience before showing it.

September 10, 2026 · 6 minRead
AI Development

Managed AI Agents: Who Saves Each Part of Your Task?

Map managed agent state across conversations, compute and business actions. Check what survives a restart with a responsibility table and recovery worksheet.

September 10, 2026 · 6 minRead
AI Development

OpenAI Agents API: What Moves Out of Your Application

OpenAI Agents API moves the agent loop into a managed runtime. Compare environment choices, recovery limits and application duties before planning a migration.

September 10, 2026 · 6 minRead
AI Development

Google Intelligent Eyewear: Gemini AI Glasses Fall 2026

Google announces Gemini-powered smart glasses with Samsung, Gentle Monster, and Warby Parker at I/O. Audio glasses ship fall 2026; display tier TBD.

May 20, 2026 · 18 minRead
AI Development

AI Search Agents Compared: Google, Perplexity, ChatGPT

Google's always-on information agents, Perplexity Pro, and ChatGPT Search compared. Which AI search agent delivers the best research results in 2026?

May 20, 2026 · 14 minRead