# Information provenance in multi-source synthesis

*Claude Certified Architect - Foundations Study Guide · Context Management & Reliability · Last updated October 22, 2026 · https://www.certifyforce.com/study-guide/claude-certified-architect-foundations/context-management-and-reliability/information-provenance-in-multi-source-synthesis*

How multi-agent research keeps every claim traceable: structured claim-source mappings with dates that synthesis merges, conflicting figures annotated rather than resolved by guesswork, established versus contested findings, format per content type, and the API's Citations and search_result blocks.

- Certification: Claude Certified Architect - Foundations
- Exam section: Context Management & Reliability
- Study guide: https://www.certifyforce.com/study-guide/claude-certified-architect-foundations

**Official exam objective:**

> Preserve information provenance and handle uncertainty in multi-source synthesis

## What the exam expects

- Explain how attribution is lost when subagent findings are compressed into prose, and require structured claim-source mappings (claim, source URL or document name, excerpt) that synthesis preserves and merges.
- Handle conflicting statistics from credible sources by keeping both values with attribution and letting the coordinator reconcile them.
- Require publication or data collection dates in structured outputs so temporal differences are not read as contradictions.
- Structure reports into well-established and contested findings that keep each source's own certainty and method, and render each content type in a fitting format.
- Use the API's Citations feature and `search_result` blocks for verifiable attribution, knowing they do not combine with `output_config.format`.

Key terms: **provenance** (where a claim came from); **claim-source mapping** (a record tying one claim to its source and supporting excerpt); **conflict annotation** (both values kept and labeled with their sources); **Citations** (the API feature that returns claims with pointers into provided documents); **`search_result` block** (a content block carrying retrieved text with a `source` and `title`).

## Key facts

| Fact | Detail | Source |
|---|---|---|
| Why attribution disappears | Findings compressed without claim-source mappings | [Exam guide](https://everpath-course-content.s3-accelerate.amazonaws.com/instructor%2F6nizmqk8tpzpfjvt6qmmav7rh%2Fpublic%2F1783542750%2FClaude+Certified+Architect+%E2%80%93+Foundations+Exam+Guide.pdf) |
| Mapping fields | Source URL, document name, relevant excerpt, plus date | [Exam guide](https://everpath-course-content.s3-accelerate.amazonaws.com/instructor%2F6nizmqk8tpzpfjvt6qmmav7rh%2Fpublic%2F1783542750%2FClaude+Certified+Architect+%E2%80%93+Foundations+Exam+Guide.pdf) |
| Conflicting credible values | Keep both, annotate with sources; coordinator reconciles | [Exam guide](https://everpath-course-content.s3-accelerate.amazonaws.com/instructor%2F6nizmqk8tpzpfjvt6qmmav7rh%2Fpublic%2F1783542750%2FClaude+Certified+Architect+%E2%80%93+Foundations+Exam+Guide.pdf) |
| Unsupported claim after drafting | Retract it if no supporting quote is found | [Reduce hallucinations](https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/reduce-hallucinations) |
| Citation reliability | API-parsed citations are guaranteed valid pointers to provided documents | [Citations](https://platform.claude.com/docs/en/build-with-claude/citations) |
| Enabling citations | All or none of the documents in a request | [Citations](https://platform.claude.com/docs/en/build-with-claude/citations) |
| Citations plus `output_config.format` | 400 error, also with deprecated `output_format` | [Citations](https://platform.claude.com/docs/en/build-with-claude/citations) |
| `cited_text` cost | Not counted toward output tokens | [Citations](https://platform.claude.com/docs/en/build-with-claude/citations) |
| `search_result` required fields | `source`, `title`, `content`; citations off by default | [Search results](https://platform.claude.com/docs/en/build-with-claude/search-results) |
| Research system citation step | A CitationAgent attributes every claim after research | [Multi-agent research system](https://www.anthropic.com/research/multi-agent-research-system) |

## Check yourself

- Why do statistics lose their sources between subagents and the final report?
- When two credible sources give different figures for the same metric, who decides how to reconcile them?
- Why can't a request use citations and `output_config.format` together?
- What must a `search_result` block include for Claude to attribute an answer to it?

## Sources

- [Claude Certified Architect – Foundations Exam Guide v1.0](https://everpath-course-content.s3-accelerate.amazonaws.com/instructor%2F6nizmqk8tpzpfjvt6qmmav7rh%2Fpublic%2F1783542750%2FClaude+Certified+Architect+%E2%80%93+Foundations+Exam+Guide.pdf)
- [Citations](https://platform.claude.com/docs/en/build-with-claude/citations)
- [Search results](https://platform.claude.com/docs/en/build-with-claude/search-results)
- [How we built our multi-agent research system](https://www.anthropic.com/research/multi-agent-research-system)
- [Reduce hallucinations](https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/reduce-hallucinations)
- [Prompting best practices](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices)

## About the full unit

The "Core concepts", "Exam traps" and "Worked scenario" sections of this unit, its 8 linked practice questions, and unlimited timed mock exams are included with the Claude Certified Architect - Foundations pack (USD $19): https://www.certifyforce.com/practice-exams