Just Now, Claude Code Evolved Again, Drafting 'Feedback Reports' for Users

marsbitОпубліковано о 2026-08-27Востаннє оновлено о 2026-08-27

Анотація

Claude Code has introduced a new feature where it can automatically draft feedback reports on behalf of the user. This occurs in scenarios such as when a tool persistently fails, Claude cannot fulfill a request, an error is identified, or the user explicitly asks for feedback. Drafts are saved locally and are not sent to Anthropic until the user reviews and approves them. Users can view, edit, or delete these drafts via a feedback queue interface and control the feature's behavior through settings. The tool is available in local, interactive terminal sessions using the Claude API but is disabled in non-interactive modes, web sessions, certain cloud platforms, and for organizations with specific data retention policies.

Claude Code has updated with another quite interesting little feature.

This time, it can help you draft feedback reports.

When a certain feature malfunctions, when Claude realizes it made a mistake, or when you point out where something went wrong, it will automatically compile the relevant situation into a report.

You can review, edit it before sending, and decide whether to approve submission.

However, you can also disable this feature in /config, or modify its behavior settings.

The official team has also compiled detailed documentation; let's take a look at the specific content.

Claude Code Starts 'Writing Feedback Reports' for Itself

Feedback drafted by Claude is a report about Claude Code written by Claude on your behalf. This feature requires using Claude Code v2.1.238 or later.

Claude Code saves each feedback draft on your local device, at the path: ~/.claude/feedback/drafts/. Nothing is sent to Anthropic until you actively send it.

Specifically, Claude will automatically draft feedback via the SendFeedback tool in the following situations:

  • A tool or command continuously fails;
  • Claude cannot complete a certain request you made;
  • You point out that Claude made a mistake, or Claude itself realizes it made an error;
  • You ask Claude to submit feedback.

Documentation address:

https://code.claude.com/docs/en/tools-reference#sendfeedback-tool-behavior

What You See When Claude Drafts Feedback

When Claude queues a feedback draft, you will see a card above the input box, displaying the title of that draft. You can:

  • Press 1 to view the draft;
  • Press 2 twice in a row to send it as-is directly;
  • Press 0 to ignore this card.

Even if you ignore the card, the draft will remain in the feedback queue.

After you ignore a feedback card, Claude Code will ask if you want to turn off Claude's automatic feedback drafting feature. If you choose not to turn it off twice in a row, it will stop asking afterwards.

By default, a maximum of 3 feedback cards are displayed per session.

After exceeding this limit, or when you set feedbackDrafts to quiet, only the number of feedback drafts currently queued will be shown at the bottom of the input box, and no specific cards will pop up.

Viewing and Editing Feedback Drafts

Run /feedback without any parameters to open the feedback queue.

This will list all unprocessed drafts from all your sessions, including those for which you closed the feedback card or never saw a card in the first place.

After selecting a draft, you can review it and perform the following operations:

Edit the title, category, and specific content;

Set Send transcript to yes or no. If the session record corresponding to when Claude created this draft still exists, this option defaults to yes, meaning the conversation segment will be submitted to Anthropic along with the feedback; choosing no sends only the feedback report itself;

Send the draft, delete the draft, or temporarily leave it in the queue to handle later.

If you want to manually write a feedback report yourself, press w to open the standard feedback window.

The /feedback command with text content, as well as the /bug command, will also directly open this standard feedback window.

Sending Feedback Drafts

When you send a draft, Claude Code submits it in the same way as a /feedback report, following the same data retention policies. After sending is complete, the draft is deleted from your local device.

If you send directly from a feedback card, it will show: ✓ Sent. If you send from the feedback queue, a receipt ID will be shown when the window closes.

Note that sending directly from a feedback card does not include the conversation transcript.

Claude Code saves your working directory information in the local draft to find the corresponding conversation transcript, but does not send this working directory.

For organizations with Zero Data Retention (ZDR) enabled, Claude Code does not provide this tool, and /feedback is similarly unavailable.

Deleting or Retaining Feedback Drafts

When you delete a draft, Claude Code simply removes it from your local device.

If you leave a draft in the feedback queue unprocessed, it will automatically expire after 30 days.

The feedback queue can save a maximum of 10 drafts across all sessions.

When Claude creates the 11th draft, Claude Code will automatically delete the earliest one.

Sessions Where Claude Cannot Automatically Draft Feedback

Claude Code only provides this feature in the following scenarios:

Interactive terminal sessions running on the user's own computer and directly using the Claude API, not through a cloud service provider.

The tool is not provided in the following scenarios:

  • Non-interactive -p run modes, and Agent SDK sessions, as these scenarios lack an interface for reviewing the feedback queue;
  • Cloud sessions, such as Claude Code on the web, because it cannot write feedback drafts to your local device;

Sessions run through the following platforms:

  • Amazon Bedrock
  • Claude Platform on AWS
  • Google Cloud Agent Platform
  • Microsoft Foundry

Sessions with the following environment variables or related configurations set:

  • CLAUDE_CODE_SEND_FEEDBACK=0
  • DISABLE_FEEDBACK_COMMAND=1
  • Setting CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC to any non-empty value
  • Feature-flag fetching is turned off;
  • Organizations that have turned off product feedback functionality;
  • Organizations with Zero Data Retention (ZDR) enabled.

This article is from the WeChat public account "Machine Heart"

Пов'язані питання

QWhat is the main new feature introduced in the latest Claude Code update according to the article?

AThe main new feature is that Claude Code can now automatically draft feedback reports for users when issues occur, such as tool failures, errors made by Claude, or user-reported problems.

QWhere are the feedback drafts saved locally on a user's device, and when is the content transmitted to Anthropic?

AThe feedback drafts are saved locally at ~/.claude/feedback/drafts/. The content is not transmitted to Anthropic until the user actively sends the report.

QHow can a user open the feedback queue to review and manage pending drafts in Claude Code?

AA user can open the feedback queue by running the /feedback command without any parameters.

QUnder what conditions will Claude automatically draft a feedback report using the SendFeedback tool?

AClaude will automatically draft feedback when: a tool or command repeatedly fails; Claude cannot complete a user's request; the user points out an error or Claude realizes it made a mistake; or the user explicitly asks Claude to submit feedback.

QIn which types of sessions is the automatic feedback drafting feature NOT available in Claude Code?

AThe feature is not available in: non-interactive -p run modes and Agent SDK sessions; cloud-based sessions like Claude Code on the web; sessions run through platforms like Amazon Bedrock, Claude Platform on AWS, Google Cloud Agent Platform, or Microsoft Foundry; and sessions with specific environment variables or configurations that disable feedback or non-essential traffic, or in organizations that have disabled product feedback or enabled Zero Data Retention.

Пов'язані матеріали

Monero Price Forecast for 2026-2032: Is it Worth Buying XMR Now?

**Article Summary: Monero (XMR) Price Forecast for 2026-2032 and Investment Outlook** This article analyzes the potential future price trajectory of privacy-focused cryptocurrency Monero (XMR). Based on technical analysis and market sentiment, it presents a bullish long-term forecast. **Key Price Predictions:** * By the end of 2026, XMR is projected to reach **$825.20**. * A maximum price of **$1,016.93** is forecast for late 2029. * The price could surge to **$1,753.23** by 2032. **Current Market Analysis (as of August 2026):** * The current price is approximately $430.66, having faced resistance at $450. * Market sentiment is labeled as "Bullish," though short-term volatility persists with support found around $430. * Technical indicators like moving averages suggest a mixed but generally positive outlook. **Investment Thesis:** Monero is presented as a compelling investment due to its core focus on transaction privacy and security, which drives demand. Its decentralized development model and growing utility are cited as foundations for long-term growth. However, the article cautions investors about significant risks, including regulatory scrutiny, market volatility, and the cryptocurrency's association with illicit activities, which could hinder mainstream adoption. **Conclusion:** While the forecast is optimistic, the article strongly advises investors to conduct thorough personal research and consult experts before investing in the volatile cryptocurrency market. The potential for XMR to reach $1,000 or surpass its all-time high is deemed possible under favorable conditions but remains highly dependent on regulatory developments and broader market dynamics.

cryptonews.ru28 хв тому

Monero Price Forecast for 2026-2032: Is it Worth Buying XMR Now?

cryptonews.ru28 хв тому

Livio, CEO of New Huo Group: RWA Will Become the Bridge Connecting Carbon-Based Humans and Silicon-Based AI, Crypto's Next 100 Million Users to Come from AI Agents

At the "Deep Integration of AI and Crypto" roundtable during the Bitcoin Asia conference, Livio Weng, CEO of Bitfire at New Huo Group, stated that the core narrative for AI and Crypto convergence in the coming years will likely focus on the "agential economy." As infrastructure for AI identity recognition, authorized wallets, and payment protocols matures, AI will progressively gain the ability to autonomously use, manage, and transact assets, deeply integrating with existing crypto sectors like trading, prediction markets, asset management, and payments. Livio believes Real World Assets (RWA) will serve as a crucial bridge connecting carbon-based humans and silicon-based AI. On one hand, RWA can tokenize traditional assets, enabling AI to use and trade assets from human society. On the other hand, "AI assets" such as computing power, data, and AI-generated services can also be made accessible to humans through RWA formats for purchase, analysis, and trading. This process will drive deeper integration between humans and AI. "The next 100 to 200 million new crypto users may not be humans, but AI Agents," Livio suggested, noting that future critical infrastructure and applications might be autonomously generated by AI rather than primarily coded by humans. As AI evolves from an information assistant to a wealth management assistant and further into an autonomous task executor, AI Agents are poised to become significant participants in the Crypto ecosystem. Livio also highlighted that deep AI-Crypto integration requires robust underlying infrastructure, including settlement layers for autonomous machine transactions, identity and authorization layers for AI independent identity and recognition, and incentive/disincentive layers to address fraud and misconduct. Protocols like X402 and EIP-8004 are exploring related applications, but challenges in security, governance, and ethics remain. In practice, New Huo is actively embracing RWA to seize opportunities from AI-Crypto convergence. The group recently launched comprehensive RWA operator services and released Hong Kong's first compliant crypto asset quantitative strategy. By incorporating RWA assets, this strategy aims to offer cross-asset allocation and risk hedging opportunities, helping investors navigate volatility across different asset classes and market cycles. Livio concluded that RWA is not only a vehicle for traditional assets to enter the on-chain world but also potential key infrastructure for AI to participate in the real economy, utilize real-world assets, and provide services to humans. As the machine economy develops, the intersection between AI and Crypto will expand, offering substantial room for innovation and growth.

marsbit35 хв тому

Livio, CEO of New Huo Group: RWA Will Become the Bridge Connecting Carbon-Based Humans and Silicon-Based AI, Crypto's Next 100 Million Users to Come from AI Agents

marsbit35 хв тому

Successes of Cheburnet: the share of free mobile internet in Central Russia decreased to 30.5%

Russia's "Sovereign Internet" project continues to advance, significantly impacting mobile internet access. In July 2026, only 30.5% of mobile data sessions in Central Russia were unrestricted, with the majority routed through a state-approved "whitelist." This represents a 31% increase in whitelist usage since January. The implementation is geographically uneven: Moscow and St. Petersburg see higher rates of free access (~49% and ~44% respectively), while border regions like Belgorod have seen unrestricted access plummet to around 12%. International forces are also contributing to digital isolation. Major certificate authority GlobalSign began revoking TLS certificates for Russian domains in June to comply with international sanctions, affecting thousands of domains, including those of major corporations like Rosneft. While its market share is small, the move aligns with the trend of separating the Russian internet segment. Domestically, authorities are preparing alternatives to maintain critical services. The government has tasked officials with ensuring the operation of key platforms like healthcare portals and payment systems during internet restrictions. The Ministry of Digital Development promotes a transition to certificates from a Russian National Certification Authority, which are natively supported by domestic browsers but may require manual installation for other devices. An AI analysis raises concerns, comparing the plan to Kazakhstan's failed 2019 attempt to mandate a state root certificate, which was blocked by major browsers as a security threat. This poses a future risk: users may eventually face a choice between foreign browsers potentially incompatible with Russian certificates or domestic software that fully trusts state-issued ones.

cryptonews.ru59 хв тому

Successes of Cheburnet: the share of free mobile internet in Central Russia decreased to 30.5%

cryptonews.ru59 хв тому

985 Professors Group Ventures into Embodied Intelligence, Raising Over ¥100 Billion in Funding This Year

In 2026, the embodied AI sector in China is witnessing a significant surge in venture capital, with individual funding rounds reaching billions of RMB. A notable trend within this boom is the rise of "academic factions," where university professors are actively founding companies. According to an IT桔子 report, over 70 embodied AI companies have founders who are or were university professors. A stricter definition identifies 18 core "academic" companies, which have collectively raised an estimated 15 billion RMB in the first eight months of 2026. These companies fall into two main categories. The first consists of firms founded by professors from top-tier universities like Tsinghua, Peking, Zhejiang, Shanghai Jiao Tong, and Fudan. Examples include Xingdong Jiyuan (陈建宇, Tsinghua), accelerating evolution (赵明国 team, Tsinghua), Galaxy Universal (王鹤, Peking University), Yunchushenchu (朱秋国, Zhejiang University), Qiongche Intelligence (卢策吾, Shanghai Jiao Tong), and Zhujidongli (张巍, Southern University of Science and Technology). The second category comprises companies incubated directly by research institutes, such as Qiuzhi Technology (from Tsinghua AIR), Delta Intelligence (from BIGAI), Xingyuan Intelligence (from Beijing智源研究院), and Zero to the Power (jointly incubated by Tsinghua). Key observations highlight Tsinghua University's dominant presence, the increasing direct investment from university-affiliated funds into these startups, and the unprecedented involvement of tenured professors. While these academic teams possess deep technical expertise in areas like robotics and control systems, they face the challenge of adapting to commercial imperatives like mass production and supply chain management. This wave represents an evolution in China's academia-to-industry model, shifting from mere patent licensing to professors leading funded ventures with institutional support, potentially accelerating the development of physical AI.

marsbit1 год тому

985 Professors Group Ventures into Embodied Intelligence, Raising Over ¥100 Billion in Funding This Year

marsbit1 год тому

Торгівля

Спот
活动图片