# Finding 084: Perplexity repeated the plain head-signal failed fetch without origin activity

## Date

2026-07-02

## Status

Published

## Summary

Perplexity was prompted in a fresh native incognito thread with
`manual-client-perplexity-plain-head-signal-repeat-20260702-001-p38`, a repeat
of the plain site-owner review prompt against the
`/lab/reading/head-signal-isolation` fixture. The prompt asked Perplexity to
open the target URL and briefly review the page, without asking for source
areas, page-head metadata, structured data, code-like tokens, marker values,
or hidden fields.

Perplexity returned `fetched:false`, `pages_opened:0`, and evidence quote
`Failed to fetch url content`. A bounded origin review found no exact
target-page hit and no related `/`, `/robots.txt`, or sitemap activity inside
the prompt window. This repeats the p37 failed-fetch answer, but differs from
p37 because the p38 window did not show ancillary PerplexityBot origin traffic.

## What does this mean?

For site owners, SEO/AEO teams, publishers, and researchers, this repeat shows that a failed AI-client fetch can appear in more than one server-side shape. In one Perplexity run the site saw root and robots traffic without a page hit; in this repeat the answer still said the page could not be fetched, but the server did not see related prompt-window origin activity at all. That makes direct server logs essential for separating page reads, site-contact-only attempts, and clean no-hit outcomes.

## Method

- Controlled-browser task:
  `research/manual-client-runs/browser-tasks/manual-client-perplexity-plain-head-signal-repeat-20260702-001-p38.browser-task.json`.
- Prompt packet:
  `research/manual-client-runs/manual-client-perplexity-plain-head-signal-repeat-20260702-001.prompts.json`.
- Response artifact:
  `research/manual-client-runs/browser-tasks/responses/manual-client-perplexity-plain-head-signal-repeat-20260702-001-p38.response.json`.
- Logged answer packet:
  `research/manual-client-runs/manual-client-perplexity-plain-head-signal-repeat-20260702-001.answers.json`.
- Prompt window:
  `2026-07-02T14:16:59Z` to `2026-07-02T14:17:38Z`.

The prompt was submitted in a fresh Perplexity native incognito thread. The UI
showed the `kai bot` account, Free plan, Search and Computer controls, an Exit
incognito control, and a generic Model selector. The high-level OpenClaw
browser `act` wrapper still returned `action targetId must match request
targetId`, so the prompt was inserted and submitted with direct Chrome
DevTools Protocol operations against the fresh visible Perplexity tab.

## Evidence

| Evidence | Result |
|---|---|
| Model answer | `fetched:false`, `pages_opened:0`; `Failed to fetch url content`. |
| Perplexity UI state | Fresh native incognito thread; visible account `kai bot`; Free plan; Search and Computer controls visible; generic Model selector. |
| Direct-origin target-page event | None for `/lab/reading/head-signal-isolation?id=manual-client-perplexity-plain-head-signal-repeat-20260702-001-p38` inside the bounded prompt window. |
| Ancillary origin activity | None for related `/`, `/robots.txt`, sitemap, or exact attempt-id paths inside the bounded prompt window. |

```json
{
  "timestampWindow": {
    "startedAt": "2026-07-02T14:16:59Z",
    "endedAt": "2026-07-02T14:17:38Z"
  },
  "confirmedHitFromPrompt": false,
  "rawEventIds": [],
  "ancillaryOriginActivity": {
    "present": false,
    "paths": [],
    "rawEventIds": []
  }
}
```

Nearby unrelated events were not counted as behavior evidence: an iPhone
Safari request for an older Perplexity finding occurred at
`2026-07-02T14:16:14.084Z`, and an `ivre-masscan` root probe occurred at
`2026-07-02T14:15:33.466Z`, both before the submitted prompt window.

## Interpretation

This p38 repeat is a clean target no-hit. It supports the narrow claim that
Perplexity again reported target-page retrieval failure under the same plain
head-signal review wording, but the server did not record PerplexityBot or
other related origin activity in the bounded prompt window.

Compared with Finding 081, the answer-side result is stable while the
origin-side trace differs. The earlier p37 run combined a failed-fetch answer
with ancillary `/robots.txt` and root requests. This p38 repeat combined a
failed-fetch answer with no related origin activity. The pair should therefore
be treated as repeated Perplexity target no-hits, not as a stable ancillary
traffic pattern.

## Limitations

- This is one Perplexity repeat from one account, plan, client mode, and time
  window.
- The prompt supplied the exact target URL, so this measures direct URL
  opening rather than independent discovery.
- The answer and origin log agree on no target-page retrieval, but they do
  not explain whether Perplexity skipped retrieval, used cached routing
  state, failed before contacting the lab, or contacted an unobserved
  intermediate service.
- The lack of related origin activity is bounded to the prompt window and the
  retained local event log.
- The high-level OpenClaw browser `act` wrapper remained unavailable for this
  tab due to the targetId mismatch, although direct CDP operations preserved
  the same fresh visible browser session.

## Publication Thesis Verification

- Thesis: In the p38 plain site-owner review repeat, Perplexity returned a
  failed-fetch answer and the lab recorded no exact target-page hit or related
  ancillary origin activity inside the bounded prompt window.
- Source: The p38 browser-task artifact, response artifact, logged answer
  packet, Perplexity UI snapshot context, and direct-origin raw event review
  for `2026-07-02T14:16:59Z` to `2026-07-02T14:17:38Z`.
- Method: Compare the copied Perplexity answer and visible browser state with
  the exact attempt id, prompt code, timestamp window, and local server-side
  events for the target path and nearby ancillary paths.
- Bias: Single-account, single-run repeat evidence from a synthetic fixture.
  The prompt is plain relative to source-area prompts but still names
  AI-assistant readability and an exact lab URL.
- Consensus: Consistent with Finding 081 on failed target-page retrieval, but
  not on ancillary origin activity. The repeated answer-side no-fetch outcome
  is stronger than any claim about stable root or robots behavior.
- Invalidation: A corrected origin review finding an exact target-page event
  or related PerplexityBot ancillary event inside the prompt window, or a
  response-artifact mismatch showing that the answer belonged to a different
  attempt, would weaken this finding.
- Verdict: Supported for the p38 run. It documents a registered clean target
  no-hit with a copied Perplexity failed-fetch answer.
- Additional tests suggested: run the prepared Copilot/Bing p38 repeat and
  then publish a p37/p38 no-hit-client comparison that separates answer
  stability from origin-side activity stability.

## Next steps

- Run the prepared Copilot/Bing p38 repeat in a fresh Temporary chat.
- Compare Perplexity p37 and p38 before claiming whether ancillary root or
  robots activity is stable for this prompt family.