PulseAugur
EN
LIVE 03:56:34

Claude Code hook fails to erase blocked prompts from transcript

By PulseAugur Editorial · [1 sources] ·

A user discovered that Claude Code's UserPromptSubmit hook, intended to block prompts from reaching the model, still logs these blocked prompts to the session transcript. Despite the hook successfully preventing the prompts from being processed by the model, six out of six tested prompts were written to the transcript, with some appearing twice. This behavior contradicts the documented functionality, which states that blocked prompts should be erased from context and not added to the transcript. AI

IMPACT This bug could impact user trust and data privacy if sensitive information is inadvertently logged despite being blocked.

RANK_REASON The item details a specific bug or unexpected behavior in a software product's feature, rather than a new release or major industry development.

Read on dev.to — Claude Code tag →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Claude Code hook fails to erase blocked prompts from transcript

How we ranked this

Signal score
25 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
The item details a specific bug or unexpected behavior in a software product's feature, rather than a new release or major industry development.
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
product, other
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
Breaking (< 6h)
Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

Full methodology in our editorial standards.

COVERAGE [1]

  1. dev.to — Claude Code tag TIER_1 English(EN) · Rulestack ·

    Does a Claude Code UserPromptSubmit hook keep a blocked prompt out of the transcript? 6 of 6 never reached the model, all 6 were still written to it

    <blockquote> <p>6 of 6 prompts that a UserPromptSubmit hook blocked never reached the model, yet all 6 were still written to the session transcript (4 of them twice), and <code>claude -p</code> exited 0 with <code>"subtype": "success"</code> every time. Context injection matched …

RELATED ENTITIES

RELATED TOPICS