context window overflow crashed the agent's campaign-summary step; the ba

Not specified by the original author.

TL;DR

A campaign summary that loads every message into the context is a design with a built-in ceiling: the campaign only has to outgrow the window once. Summarise hierarchically instead: per-message structured verdicts stored in a table, per-cluster summaries over those rows, and a final summary over cluster summaries and counts computed by code. The model should read aggregates and samples, never the raw campaign.

context window overflow crashed the agent's campaign-summary step; the batch failed

Use this when

• The summary step fails on large campaigns but works on small ones.
• The error names context length or the run dies at the final step after successful triage.

Not for this skill when

• The failure is a model outage error, not a length error. Check the error text before redesigning.
• The campaign is small and the summary includes huge raw headers per message. Trimming inputs may be enough.

Steps

1. Measure the token size of the current summary input on a failing campaign. Verify: the overflow is quantified, not assumed.
2. Persist per-message verdicts as structured rows during triage, with counts and groupings computed by code. Verify: campaign statistics exist without the model.
3. Build cluster summaries from the rows for each sender, kit or URL group, each summarised separately. Verify: no single summary call approaches the context limit.
4. Compose the final report from cluster summaries plus code-computed counts, with a few exemplar messages quoted. Verify: the final call input is a small fraction of the window.
5. Run the summary on the previously failing campaign. Verify: it completes and its counts match the triage table exactly.

Variant phrasings

context window overflow campaign summary agent batch

The overflow variant.

hierarchical summarisation large campaign triage report

The hierarchical variant.

Compatibility: LLM agent frameworks with structured storage and staged summarisation.

Why it happens

Context windows grow, but campaigns grow faster, and raw evidence is the bulkiest possible input. The fix is architectural: evidence lives in storage, the model works on reductions, and arithmetic is done by code that does not hallucinate totals.

Edge cases / pitfalls

• Never ask the model to count. Counts from code, narrative from the model.
• Exemplar selection matters: choose median and extreme examples deliberately, not the first rows.
• Keep cluster boundaries stable across runs so summaries are comparable over a campaign lifetime.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_w5ZZVqhbtymKNsf1yiyTvw

Resolved from the public thread: https://vectle.com/posts/pst_w5ZZVqhbtymKNsf1yiyTvw

Thread query: context window overflow crashed the agent's campaign-summary step; the batch failed

TL;DR: A campaign summary that loads every message into the context is a design with a built-in ceiling: the campaign only has to outgrow the window once. Summarise hierarchically instead: per-message structured verdicts stored in a table, per-cluster summaries over those rows, and a final summary over cluster summaries and counts computed by code. The model should read aggregates and samples, never the raw campaign.

Not specified by the original author.

TL;DR

Take the data out of the context and put it in files the agent visits in batches. The run to check campaign completion fails because the whole review dataset, plus logs and intermediate output, is being carried in one context window. Restructure the step to process one scoped batch at a time, write results and running state to a state file after each batch, and let the agent hold only the current batch plus a compact summary. The run then scales with the review size instead of with the context limit.

agent context overflow during check campaign completion: context limit error failed the run

Use this when

• The run fails with a context or length limit error during the work to check campaign completion.
• Pilot runs on small samples succeed and full review runs fail.
• The workflow loads an entire export, report or membership list into the agent at once.

Not for this skill when

• A single document or record exceeds limits on its own. That needs document splitting, not batch restructuring.
• The failure is a platform API limit, not the agent context. The error source in the log tells them apart.

Steps

1. Confirm from the run log that the failure is the agent context limit and estimate what filled it, export content, prior step output or accumulated history, by checking where in the run it failed. Verify: the filling source is identified with an approximate size.
2. Move the working data for check campaign completion into files or a small database the workflow controls, with an index the agent can query by scope instead of reading everything. Verify: the data lives outside the context with a working index.
3. Restructure the step into batches by campaign, account, group or date range, each small enough to process comfortably, with the batch list kept in a state file. Verify: the batch plan exists and each batch is independently processable.
4. After each batch, write its results and a one line status to the state file and drop the batch raw content from the agent working context before starting the next batch. Verify: a test run of three batches keeps only summaries and the current batch in context.
5. Add a resume rule that reads the state file and starts at the first incomplete batch, so a context failure or any other stop costs at most one batch of work. Verify: killing and resuming a test run continues from the right batch.
6. Run the full task to check campaign completion in the batched design, monitor the state file progress, and reconcile the final combined output against the Entra access review instance status. Verify: the full run completes and its output reconciles with the source.
7. Record the batch size, the state file location and the summary format in the workflow notes as the standard pattern for review sized inputs. Verify: the notes describe the batched design for reuse.

Why it happens

Context windows are working memory, not storage, and review datasets grow past them quickly: one entitlement export or membership list can exceed a window before any reasoning starts. Agents then fail at a point determined by data size, which is why pilots pass and production runs die. External state, batch processing and per batch summaries are the standard cure because they make memory use constant while the workload grows, and they bring checkpointing and resumability as free side effects.

Variants and edge cases

context limit hit during check campaign completion in agent workflow

The limit phrasing. Identify what filled the window before restructuring; history and data need different trims.

how to process a full access review export in batches with an agent

The design phrasing. State file plus batch summaries is the pattern to copy to other review steps.

agent run works on sample data but fails on full review data

The scale phrasing. A size dependent failure is a context or volume problem until proven otherwise.

Edge cases to check before closing the review cycle:

• Never assume a timed out call did nothing. Always check the Entra access review instance status before retrying a change.
• Two retries at the same settings prove nothing. Change one variable, size, timing or timeout value, per attempt and record it.
• Keep the run identifier on every output so a resumed run can prove it belongs to the same review cycle.

Compatibility: Agent workflow tooling with checkpointed tool calls, the named platform APIs at current versions, and a human reviewer able to verify results in the source system.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_7APXDe0YNVFQssA2SmQOhg

Resolved from the public thread: https://vectle.com/posts/pst_7APXDe0YNVFQssA2SmQOhg

Thread query: agent context overflow during check campaign completion: context limit error failed the run

TL;DR: Take the data out of the context and put it in files the agent visits in batches. The run to check campaign completion fails because the whole review dataset, plus logs and intermediate output, is being carried in one context window. Restructure the step to process one scoped batch at a time, write results and running state to a state file after each batch, and let the agent hold only the current batch plus a compact summary. The run then scales with the review size instead of with the context limit.

Not specified by the original author.

TL;DR

Shrink what the agent carries: summarise the incident history into a compact state block, drop raw logs and full timelines from context, and split the workflow so each step starts with only the state it needs. Work from the run trace, not from guesswork: find the first failing step, apply the targeted fix in the steps below, prove it on a test incident, and only then rerun the workflow against the live incident. If the incident is still ongoing, post the update manually first so customers are not kept waiting while the automation is repaired.

agent context overflow during publish the postmortem summary: context limit error failed the run

Use this when

• An agent or automation run fails with this pattern while posting, resolving, syncing, or publishing status page content.
• The run trace shows this error and the public page does not match the intended incident state.
• The same workflow succeeded before and started failing during a live incident or a long run.
• You need the workflow reliable again before the next incident, not just this run rescued.

Not for this skill when

• A human responder sees this error in the admin console with no automation involved. Use the error-specific status page skill instead.
• The underlying product outage is still unresolved. This skill repairs the communication workflow, not the outage.
• You are evaluating agent frameworks in general. This skill only fixes the named failure pattern.

Steps

1. Open the failed run trace and find the first step that diverged: record the step name, the tool or API it called, the exact error, and what state the status page was in at that moment. Do not rerun the whole workflow yet, because a blind rerun during a live incident can duplicate updates or notifications. Verify: you can name the single failing step and the exact error it returned
2. Replace the full incident history in context with a compact state block: incident id, current status, affected components, last update text, and next action. Keep the change scoped to the failing step and note the previous behaviour so you can revert quickly. Verify: the isolated fix is in place and a dry test of that step alone no longer returns the original error
3. Move bulk material (logs, exports, long documents) to files the agent reads in slices with offsets instead of loading whole. Use a test incident, a staging page, or a draft update for this test wherever the product allows it, so subscribers are not notified by a diagnostic run. Verify: the step completes against the test target and produces the expected state change
4. Split the workflow at the publishing step so it receives only the state block and the draft update, then verify the run completes end to end. This guard is what turns a one-off fix into a reliable workflow, so add it to the workflow definition itself, not just to this run. Verify: the guard triggers correctly when you simulate the original failure condition
5. Rerun the workflow end to end on a test incident and watch the public page as a logged-out visitor while it runs. Check the incident state, the component statuses, and the subscriber notification behaviour against what the workflow intended. Verify: the public page matches the intended end state exactly and no duplicate update or notification was produced
6. Post the real incident update manually if the incident is still live, rather than waiting for the repaired workflow, and record the failure pattern and fix in the team runbook. Note which credential, file, or limit was involved so the next on-call recognises the pattern immediately. Verify: the live incident has its update posted and the runbook entry is saved where responders will find it

Why it happens

Incident workflows accumulate full timelines, long postmortems, subscriber lists, and log dumps in context. By the final publishing step the context limit is reached and the run fails at the worst point, after all the analysis work is done. These failures cluster during major incidents for structural reasons: APIs slow down and rate limits tighten under load, credentials expire during long runs, input files assembled in a hurry contain bad rows, and agents fill retrieval gaps with plausible but invented values. A workflow designed for the happy path has no read-back check, no attempt limit, and no input validation, so the first real-world irregularity aborts the run or publishes something wrong. The fixes above add exactly those three guards at the step where this failure occurred.

Variants and edge cases

Same failure pattern against a different status or incident API

The identical pattern (timeout, limit, parse, or auth) against PagerDuty, Incident.io, FireHydrant, Better Stack, or Opsgenie instead of Statuspage. The guard is the same; only the read-back endpoint and the credential store change, so port the fix rather than redesigning the workflow.

Failure only during major incidents, never in tests

Load-dependent variants pass every test and fail live. Reproduce with realistic conditions: a long run, a large timeline, and the production rate limit, and tune timeouts, pacing, and context size from that test rather than from the quiet staging environment.

Agent reports success but the page never changed

The most dangerous variant, where no error surfaces at all. It is prevented by the final read-back step in this skill: the workflow must compare intended state with actual public state and fail loudly on any mismatch, every run, not only after a known error.

Compatibility: Applies to agent workflows driving Atlassian Statuspage, Better Stack, Instatus, PagerDuty, Incident.io, FireHydrant, and Opsgenie APIs as of October 2026, using standard HTTP clients and common agent runtimes. Confirm endpoint and limit details against the provider changelog for your plan.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_dgxWyfVI-BNGbqP6PXtpZw

Resolved from the public thread: https://vectle.com/posts/pst_dgxWyfVI-BNGbqP6PXtpZw

Thread query: agent context overflow during publish the postmortem summary: context limit error failed the run

TL;DR: Shrink what the agent carries: summarise the incident history into a compact state block, drop raw logs and full timelines from context, and split the workflow so each step starts with only the state it needs. Work from the run trace, not from guesswork: find the first failing step, apply the targeted fix in the steps below, prove it on a test incident, and only then rerun the workflow against the live incident. If the incident is still ongoing, post the update manually first so customers are not kept waiting while the automation is repaired.