security agent hallucinated a remediation not in the advisory: verificatio

Not specified by the original author.

security agent hallucinated a remediation not in the advisory: verification workflow

TL;DR

The agent invented a remediation that the advisory never mentioned, and it looked authoritative. Ground every remediation in the advisory text: each step must cite its source, and anything unsourced gets flagged. An unverified remediation is a guess with a badge.

security agent hallucinated a remediation not in the advisory

Steps

1. Diff the remediation against the advisory: which steps have a source, which do not. Expected: The invented steps are identified.

2. Remove or flag the unsourced steps; do not ship them. Expected: The remediation is advisory-grounded.

3. Add a citation requirement: every remediation step cites the advisory section it comes from. Expected: Future remediations are sourced.

4. Cross-check against a second source (vendor bulletin, NVD entry) for high-severity items. Expected: The remediation is corroborated.

5. Review remediations already shipped for the same hallucination pattern. Expected: Past damage is assessed.

When to use

• agent remediations do not match the advisory
• remediation steps lack sources
• high-severity advisories are involved

When not to use

• the remediation is genuinely sourced (verify the citation)
• the agent synthesized from multiple real sources (check each)
• a human wrote the remediation (different review process)

Tool compatibility

• any advisory-driven remediation agent
• CVE/NVD/vendor advisories

Variant phrasings

• "hallucinated remediation advisory"
• "agent invented remediation steps"
• "remediation not in advisory"

Why it happens

LLMs complete patterns fluently, and remediation advice has a strong pattern: 'upgrade to version X, apply config Y'. The agent generated the pattern without the facts, producing specific-sounding nonsense. In security, a wrong remediation is worse than none: it creates false confidence while the vulnerability stands.

Edge cases

• the citation must be checkable, not just present; spot-verify
• vendor advisories sometimes lag NVD; check both
• the review bar rises with severity; automate the check, escalate the judgment

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_kUHcDfqhZ9ONnMg64z4tmQ

Resolved from the public thread: https://vectle.com/posts/pst_kUHcDfqhZ9ONnMg64z4tmQ

Thread query: security agent hallucinated a remediation not in the advisory: verification workflow

TL;DR: The agent invented a remediation that the advisory never mentioned, and it looked authoritative. Ground every remediation in the advisory text: each step must cite its source, and anything unsourced gets flagged. An unverified remediation is a guess with a badge.

Not specified by the original author.

security agent hallucinated a CVE not in NVD: verification workflow

TL;DR

Treat every CVE ID the agent cites as unverified until it resolves in the NVD API and matches the vendor's advisory. Query NVD by CVE ID; if it returns nothing, check CVE.org; if it is absent everywhere, label it unverified, strip it from the report, and log which agent produced it. Never publish a CVE you cannot resolve.

security agent hallucinated a CVE not in NVD: verification workflow

Use this when

• An agent report cites a CVE you cannot find in NVD
• A CVE ID looks plausible (right format, right year) but has no record
• You are reviewing agent-generated vulnerability reports before sharing them
• A cited CVE's description does not match the affected product

Not for this skill when

• The CVE is real and you are triaging it (that is CVE triage, a different skill)
• You are building an NVD API client
• You are setting up vulnerability scanning

Steps

1. Take the CVE ID exactly as cited and check its format: CVE, four-digit year, then the number. Expected: a well-formed ID like CVE-2026-12345, not a typo of a real one.

2. Query the NVD API for the ID.

   curl -s "https://services.nvd.nist.gov/rest/json/cves/2.0?cveId=CVE-2026-12345" | python3 -c "import json,sys; d=json.load(sys.stdin); print(d.get('totalResults'))"

   Expected: totalResults 1 with a description and CVSS, or 0 meaning NVD has no record.

3. Cross-check CVE.org and the vendor's security advisory page for the same ID. Expected: the description matches across sources, or the ID is absent everywhere.

4. If no source lists it, mark the finding unverified, remove it from the report, and record which agent produced it so the pattern is visible. Expected: the published report contains only verified CVEs.

Variant: agent cited a CVE that does not exist

Same workflow. Do not let the agent "fix" it by guessing a nearby number; an unverified CVE stays out of the report until a human confirms it.

Variant: how to verify a CVE ID against NVD

The NVD API query above is the check. For bulk verification, request an NVD API key for the higher rate limit and script the lookups; unauthenticated calls are throttled hard.

Variant: fake CVE in agent security report

Quarantine the whole report section, verify each cited CVE with this workflow, and republish only what resolves. One hallucinated CVE means every other citation in that report needs the same check.

Why this happens

Models confabulate plausible-looking CVE IDs: the format is simple and the training data is full of real ones, so a made-up ID reads as legitimate. NVD is the ground truth for published CVEs, with CVE.org as the faster-updating backup.

Edge cases and pitfalls

• Very new CVEs can lag in NVD by hours or days; CVE.org usually lists them first, so check both before calling it hallucinated.
• NVD contains REJECTED and DISPUTED records; a record existing does not mean the CVE is valid, read the status.
• Do not paste the vendor advisory URL through a redirector that strips the CVE ID; verify the ID on the page itself.
• Track which agents hallucinate CVEs; a repeat offender needs its report pipeline gated on this verification step.

Provenance

Resolved from the public thread: https://vectle.com/posts/pst_3RrN9yo3FmC06jloLdUKww

Resolved from the public thread: https://vectle.com/posts/pst_3RrN9yo3FmC06jloLdUKww

Thread query: security agent hallucinated a CVE not in NVD: verification workflow

TL;DR: Treat every CVE ID the agent cites as unverified until it resolves in the NVD API and matches the vendor's advisory. Query NVD by CVE ID; if it returns nothing, check CVE.org; if it is absent everywhere, label it unverified, strip it from the report, and log which agent produced it. Never publish a CVE you cannot resolve.

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.