Where this comes from
Primary sources
- Effective harnesses for long-running agents · Anthropic (accessed 09/20/2026)
Thread
What the model sees
One question followed across the site: what is in front of the model this turn, and who decided to put it there. The answers run from a written instruction to a note a session leaves for the next one.
The organizing sentenceA model knows what it was trained on and what is in this request; everything else is a choice somebody made.
Stored somewhere is not the same as seen now.
The model works from the context actually supplied to this request. These are different ways to assemble it.
Context for this call
- 1Instructions + current input
- 2Select and fit
- 3Model request
Choose relevant evidence and instructions within the available context.
BoundaryMore context can add distractors and cost.
Memory across calls
- 1Selected prior information
- 2Store, update, retrieve
- 3Current context
Persist useful facts or state, then retrieve what this request needs.
BoundaryStored memory can be stale, wrong, or inappropriate to retain.
Work across sessions
- 1Task state + artifacts
- 2Checkpoint and resume
- 3Next work session
Carry progress, blockers, and inspectable artifacts into the next session.
BoundaryA summary alone is not an action receipt or proof the task finished.
A focused business & team operations example. Additional perspectives appear where they provide a useful contrast.
GUIDED WORKED EXAMPLE Fictional fixtures · scripted outputs · no live model or external actions
What the model sees: see it in practice.
Tracing which information reaches each model request and why it was included.
What you’ll walk through
Follow the information available at a particular step and compare it with the wider task history. Inspect what changes after truncation, retrieval, or summarization.
The task in this version
Resume an investigation after conversation summarization.
What you’ll learn to check
Before/after context trays, omitted evidence, restored constraints, and a corrected supported answer.
The setting makes the example concrete. Carry the underlying pattern into your own work; adapt the sources, tools, and level of oversight to your task.
Business & team operationsAn authored case with its own evidence, changed condition, and decision.
Resume an investigation after conversation summarization.
Authored case. Select any record below; nothing is sent to a model.FOLLOW THE EXAMPLE1 / 6
Interpret this honestlySample evidence, not your actual data.No real messages, tools, training, or hardware operations run.The sequence illustrates the concept; it is not a recorded agent trace.
THE VISIBLE WORKStarting evidence
Input record
AUTHORED TEACHING RECORD · NOT A LIVE RUN
Summary: warranty question pending. Omitted: use manual v3, not v1. Both files exist.
What changed: Establish the facts supplied for this version of the task.
WHY THIS MATTERS
What this case assumes
The model cannot rely on details that are absent from its current input unless another mechanism retrieves them. A summary may drop unresolved conditions.
1 / 6
Apply this to your project
Describe your task to your own model and use What the model sees as a reference. Ask whether it fits, which alternatives meet the same automation needs, and how you would implement and check the result.
For a normal inference request, a language model works from learned parameters and the supplied context. Files, past conversations, and tool results become available through the surrounding system.
Ask what is stored, who selects it, and what is actually included now.
Choose information at the right boundary
| Mechanism | What it contributes | What you still decide |
|---|---|---|
| Prompting | Instructions, examples, output contract | What success means |
| Context assembly | Material supplied for this call | Relevance, order, access, budget |
| Retrieval | Selected material from a collection | Search, source identity, evidence coverage |
| Memory | Information retained for later use | What to store, update, expire, or delete |
| Skills | Reusable procedures and resources | When to load them and their authority |
| Checkpoints | State across sessions | What is durable and how to resume |
Four distinctions to keep straight
Retrieval selects material, not necessarily truth. Sources can contain instructions, errors, obsolete policy, or malicious text. Distinguish source material from instructions that govern the task.
A summary can lose a qualification. Keep a route back to original records when omitted detail could change the answer. Summaries and structured records can coexist.
Memory is useful within sessions too. A long conversation can use external state. Applications can save task records deterministically, and agents at several levels can select what to remember. Model-directed memory is not exclusive to level 7.
A note is not the whole environment. Resuming work can require files, tool state, action receipts, permissions, and unfinished-task records. Restore and inspect them before relying on the last summary.
Try a revealing case
Supply a current policy and a conflicting remembered fact. Does the system notice the conflict and preserve provenance? Remove the relevant source. Does it admit the gap?
Relevant, sufficient context is the target. More tokens can add cost and distractions; a large context window does not guarantee useful attention to every fact. Caching behavior and pricing vary by provider.
Last reviewed 09/20/2026. Pages unreviewed for 90 days are flagged for another pass.