# Finding 010: Perplexity Controlled-Browser Directive-Noindex No-Hit

## Status

Confirmed no-hit (bounded evidence window)

## Summary

Perplexity (incognito mode, free/basic search tier) was prompted with a
directive-noindex compliance test (`directive-noindex`, prompt code
`manual-client-perplexity-20260625-001-p06`) via a fresh Perplexity chat
in the OpenClaw controlled browser profile. Perplexity returned
`fetched:false`, `pages_opened:0`, `confidence:0.19`, stating the
available browsing capability was not usable for opening the target URL.
No matching direct-origin server events were found in the bounded
`2026-06-28T01:46:08Z–01:46:45Z` UTC window.

## Test Details

- **Attempt ID:** `manual-client-perplexity-20260625-001-p06`
- **Source Prompt ID:** `directive-noindex`
- **Prompt Code:** `manual-client-perplexity-20260625-001-p06`
- **AI System:** Perplexity
- **Surface:** perplexity-web (incognito mode)
- **Triggered By:** openclaw-controlled-browser
- **Target URL:** `https://ai-crawler-lab.kaistone.ai/lab/directives/meta-noindex?id=manual-client-perplexity-20260625-001-p06`
- **Timestamp Window:** `2026-06-28T01:46:08.000Z` – `2026-06-28T01:46:45.000Z` UTC

## Model Response

```json
{
  "attempt_id": "manual-client-perplexity-20260625-001-p06",
  "source_prompt_id": "directive-noindex",
  "prompt_code": "manual-client-perplexity-20260625-001-p06",
  "ai_system": "Perplexity",
  "fetched": false,
  "pages_opened": 0,
  "answer": "I could not open the target page in this turn, so I cannot verify its robots/meta directive directly.",
  "evidence_quote": "Since a user requested the fetch, this fetcher generally ignores robots.txt rules.",
  "limitations": "The available browsing capability in this turn was not usable for opening the target URL, so no page content or meta directives were observed from the target page itself.",
  "confidence": 0.19
}
```

## Origin Evidence Review

- **Raw Event IDs:** none
- **Server Page Count:** 0
- **Resource Fetch Count:** 0
- **Client Capability Count:** 0
- **IP Matches:** none
- **Confirmation Status:** `registered_no_hit`
- **Confirmed Hit From Prompt:** false

The bounded evidence window (`01:46:08–01:46:45 UTC`) contains zero
direct-origin events matching this attempt. Consistent with all prior
Perplexity controlled-browser results.

## Notable Observation

Perplexity's response included the quote "Since a user requested the
fetch, this fetcher generally ignores robots.txt rules" — this is a
policy-level claim about Perplexity's fetcher behavior. However, since
Perplexity did not actually fetch the target page (no direct-origin
evidence), this claim is unverifiable from this test alone. The low
confidence (0.19) and the statement that "browsing capability was not
usable" suggest that Perplexity's tool layer was again restricted in
incognito mode, consistent with the p05 observation.

## Publication Thesis Verification

- Thesis: Perplexity web chat (incognito, free/basic tier) does not fetch
  target URLs from the lab server when prompted with a directive-noindex
  compliance test, producing no direct-origin server events.
- Source: Direct-origin server logs from lab infrastructure.
- Method: Bounded timestamp window matching with attempt_id correlation.
- Bias: Negative result reduces confirmation bias.
- Consensus: Consistent with all prior Perplexity no-hit results (p01–p05,
  p09, p15–p19).
- Invalidation check: No evidence found that would invalidate the no-hit
  conclusion. The policy claim about ignoring robots.txt is noted but
  untestable since no fetch occurred.
- Verdict: `supported_for_recorded_attempt`
- Confidence: high for the recorded attempt.
- Additional tests suggested: re-run with Perplexity Pro tier; test with
  non-incognito Perplexity chat to isolate whether incognito mode is the
  causal factor; explore whether the robots-ignoring claim can be verified
  when Perplexity does successfully fetch a page.