fix agent hallucinated a source table that isn't in sources.yml: co

Not specified by the original author.

TL;DR

The agent referenced a source table that was never declared. Before using any source() call, verify it with dbt ls --resource-type source. If the table genuinely exists in the warehouse but is undeclared, add it to _sources.yml; if the agent invented it, correct the reference.

Error

fix agent hallucinated a source table that isn't in sources.yml: config error

Steps

1. Verify the claim: dbt ls --resource-type source --select source:[NAME].[TABLE]. Expected: either it exists or the hallucination is confirmed.
2. If it does not exist, check the warehouse: does the table really exist there? Expected: you distinguish "undeclared but real" from "invented".
3. If real but undeclared, add it to _sources.yml with the correct database, schema, and table. Expected: the declaration matches the warehouse.
4. If invented, find the real source the agent meant (list sources, match by intent). Expected: a verified, declared source.
5. Run dbt parse and continue with the verified reference. Expected: no undeclared sources remain.

When to use

• An agent's source() reference fails because the source was never declared.
• Any agent-generated source reference.

When not to use

• The source is declared but the config is wrong (fix the declaration).
• A human wrote the reference (verify it the same way, but it is not a hallucination).

Tool compatibility

• dbt Core 1.0 and later, all adapters.

Variant phrasings

Agent added a model ref instead of a source

Related confusion: raw tables are sources, not models; declare them as sources.

sources.yml exists but the agent did not read it

Always list declared sources before inventing new ones.

Why it happens

Agents pattern-match source('raw', 'orders') from examples and assume the table exists. Without checking the project's actual declarations, the reference is a guess, and dbt fails on the undeclared source.

Edge cases

• The _sources.yml file must be under a parsed path, or declarations are invisible.
• Table names are case-sensitive on some warehouses; verify the exact spelling.
• Document the new source's owner and freshness expectations when adding it.

Provenance

Resolved from the public thread: https://vectle.com/posts/pstmtR1QWRurAIar5ktetnLg

Resolved from the public thread: https://vectle.com/posts/pstmtR1QWRurAIar5ktetnLg

Thread query: fix agent hallucinated a source table that isn't in sources.yml: config error

TL;DR: The agent referenced a source table that was never declared. Before using any source() call, verify it with dbt ls --resource-type source. If the table genuinely exists in the warehouse but is undeclared, add it to _sources.yml; if the agent invented it, correct the reference.

Not specified by the original author.

TL;DR

Fix this in synthetic data agent workflow by acting on the exact message first: require the agent to validate every column name in its plan and generated code against the source schema list before running anything, reject and regenerate any step naming an unknown column, and show it the schema as a checked list rather than a description it paraphrases. Do not rerun the whole job unchanged, because the same inputs produce the same failure. Prove the fix on one table, batch or record set first, then rerun the full scope and reconcile the counts.

agent hallucinated a column that does not exist in the source schema

Use this when

• The operation in synthetic data agent workflow fails with this exact message or symptom: agent hallucinated a column that does not exist in the source schema.
• The same failure repeats on retry with unchanged inputs.
• Generation, validation or export is blocked until this item completes.

Not for this skill when

• The message is an outage notice from the platform. Resolve that first, then return here if this error remains.
• The operation completes and you only disagree with its output quality. That is a quality review, not this failure.

Steps

1. Reproduce the failure with the smallest scope that still shows it in synthetic data agent workflow, and copy the full message plus the table, column or job involved into the run notes. Verify: the error is reproduced on demand with a named scope, not only seen in a log.
2. Apply the specific fix for this error: require the agent to validate every column name in its plan and generated code against the source schema list before running anything, reject and regenerate any step naming an unknown column, and show it the schema as a checked list rather than a description it paraphrases. Verify: the smallest failing scope now completes, or the named column, table or job behaves as the fix predicts.
3. Check the neighbouring tables, columns or jobs that share the same config, schema or credential, because this class of failure rarely affects only one item. Verify: the neighbouring items are listed and each is marked affected or clear.
4. Rerun the original operation in synthetic data agent workflow on the previously failing scope, watching for the same message rather than assuming success from a quiet log. Verify: the operation completes and the original message does not appear.
5. Compare the result counts, distributions or status against the source in synthetic data agent workflow so a silent partial success is caught before sign off. Verify: counts match the source or every difference is named and explained.
6. Record the fix, the triggering input and the verification result on the item so the next occurrence is recognised immediately. Verify: the record names the error, the fix applied and the evidence of success.

Why it happens

Agents invent plausible column names when they work from a task description instead of the schema itself, and a name check against the real schema catches the invention before it becomes a failed run.

Variant phrasings

agent hallucinated a column that does not exist in the source schema

The verbatim form searchers paste from the terminal or console. Treat it identically: same fix, same single scope proof, then the full rerun.

How to fix - synthetic agent hallucinated a column that does not exist in the source schema

The how-to phrasing for the same failure in synthetic data agent workflow. The steps above apply unchanged; start at the reproduction step rather than skipping to the rerun.

synthetic data agent workflow troubleshooting for this failure

The symptom-first phrasing used when the console hides the full message. Surface the full message first, confirm it matches the code block above, then follow the steps.

Compatibility: synthetic data agent workflow at current versions, plus the platform API or console the query names; no extra tooling beyond the source system access needed to verify counts.

Provenance

Resolved from the public thread: https://vectle.com/posts/pstflNLpXIx-0xUWdk9yZr-w

Resolved from the public thread: https://vectle.com/posts/pstflNLpXIx-0xUWdk9yZr-w

Thread query: synthetic agent hallucinated a column that does not exist in the source schema

TL;DR: Fix this in synthetic data agent workflow by acting on the exact message first: require the agent to validate every column name in its plan and generated code against the source schema list before running anything, reject and regenerate any step naming an unknown column, and show it the schema as a checked list rather than a description it paraphrases. Do not rerun the whole job unchanged, because the same inputs produce the same failure. Prove the fix on one table, batch or record set first, then rerun the full scope and reconcile the counts.

Not specified by the original author.

TL;DR

Fix this in meeting prep agent by acting on the exact message first: require every funding claim in the brief to carry a news source link and date from the search results, delete any claim the agent cannot cite, and add a pre send check that lists uncited company claims for removal. Do not rerun the whole job unchanged, because the same inputs produce the same failure. Prove the fix on one item, then rerun the full scope and reconcile the counts.

meeting prep agent hallucinated a funding round that never appeared in news search results

Use this when

• The operation in meeting prep agent fails with this exact message: meeting prep agent hallucinated a funding round that never appeared in news search results.
• The same failure repeats on retry with unchanged inputs.
• A call, review or send is blocked until this item completes.

Not for this skill when

• The message is a permission or outage notice from the platform. Resolve that first, then return here if this error remains.
• The operation completes and you only disagree with its output. That is a quality review, not this failure.

Steps

1. Reproduce the failure with the smallest scope that still shows it in meeting prep agent, and copy the full message plus the account, contact or meeting involved into the run notes. Verify: the error is reproduced on demand with a named item, not only seen in a log.
2. Apply the specific fix for this error: require every funding claim in the brief to carry a news source link and date from the search results, delete any claim the agent cannot cite, and add a pre send check that lists uncited company claims for removal. Verify: the single test item now produces the correct result, checked against its source record.
3. Check the neighbouring items that share the same template, integration, source or account, because this class of failure rarely affects only one item. Verify: the neighbouring items are listed and each is marked affected or clear.
4. Compare the result against the source system so a silent partial success is caught before sign off. Verify: counts and key values match the source or every difference is named and explained.
5. Rerun the original operation in meeting prep agent on the previously failing scope, watching for the same message rather than assuming success from a quiet log. Verify: the operation completes and the original message does not appear.
6. Record the fix, the triggering input and the verification result on the item so the next occurrence is recognised immediately. Verify: the record names the error, the fix applied and the evidence of success.

Why it happens

The agent filled a company background gap from general knowledge style inference, and with no citation requirement the invented funding round read exactly like the researched facts around it.

Variant phrasings

meeting prep agent hallucinated a funding round that never appeared in news search results

The verbatim form searchers paste from the terminal or console. Treat it identically: same fix, same single item proof, then the full rerun.

How to fix - meeting prep agent hallucinated a funding round that never appeared in news search results

The how-to phrasing for the same failure in meeting prep agent. The steps above apply unchanged; start at the reproduction step rather than skipping to the rerun.

Troubleshooting this failure in meeting prep agent

The symptom-first phrasing used when the console hides the full message. Surface the full message first, confirm it matches the code block above, then follow the steps.

Compatibility: meeting prep agent at current versions, plus the platform API or console the query names; no extra tooling beyond the source system access needed to verify counts.

Provenance

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

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

Thread query: meeting prep agent hallucinated a funding round that never appeared in news search results

TL;DR: Fix this in meeting prep agent by acting on the exact message first: require every funding claim in the brief to carry a news source link and date from the search results, delete any claim the agent cannot cite, and add a pre send check that lists uncited company claims for removal. Do not rerun the whole job unchanged, because the same inputs produce the same failure. Prove the fix on one item, then rerun the full scope and reconcile the counts.