You can find them linked [1] in the OG article from Anthropic [2].
[1] https://www.mozilla.org/en-US/security/advisories/mfsa2026-1...
170 karma · joined June 13, 2022
[1] https://www.mozilla.org/en-US/security/advisories/mfsa2026-1...
@moltbook is a cool social experiment, but its trust model will likely undermine its long-term sustainability.
An agent is deemed trustworthy simply by (1) requesting an API key with its name and description, then (2) having a human validate a claim URL by (3) posting an account-specific code on X.
The platform relies on moderators (probably just @MattPRD and some friends) to stop malicious agents or humans spreading prompt injections for secret leaking, hysteria-inducing robo-apocalyptic posts, and crypto scams. This is a huge opportunity for frontier labs or inference providers to implement prompt attestation, the equivalent of digital signatures:
1. The agent calls an inference API to generate text to be posted on Moltbook. 2. The inference API, hosted by Anthropic, OpenAI, X AI, OpenRouter, or others, receives the prompt. 3. The model generates the output. 4. The output is sent back to the user with a prompt signature over the input prompt and the generated output. 5. The agent posts the message along with the prompt signature. 6. Other agents or humans can check the legitimacy of the agent and its post by sending the post content and the prompt signature to a validation API hosted by the inference API provider.
Remote attestation is the alternative: validating that trusted binaries such as Claude Code or OpenClaw were indeed running on a trusted platform and are the ones generating the output and posting to Moltbook. This is a familiar struggle for companies behind mobile operating systems, social media platforms, and games. The task is Sisyphean: the device remains under the user's control, and dynamic instrumentation has become commoditized.
Why not have all the articles in the new issue directly piped into your favorite RSS reader?
iosifache[.]me/feeds/phrack.xml
Thank you so much for taking the time to explain!
> Their copyright holder may step forward at any time after you've treated it as unlicensed.
Does this mean that Satoshi can just come and claim that a whole industry is using his/her copyrighted material?
Following a suggestion from @irskep, I've added CLI command support for the search and download features.
This raised a valid concern - while we're focused on building MCP servers, we shouldn't overlook whether users already have preferred (T/G)UIs available. When they don't exist, we should consider user experience and make our functionality accessible through multiple interfaces beyond just MCP.
I justified the hours I invested by thinking I could search, download, and explore books directly from Claude Desktop. While the initial steps are achievable with a CLI tool, the integration opens up new possibilities.
Some general thoughts:
- You’ll find the MCP mental model similar to the API one. - MCP integrations make it easier for non-technical users to access tools that were previously too technical. - An MCP integration implicitly respects a contract, unlike CLIs and GUIs which involve human aspects (aesthetics, information organisation, etc.). - MCP is an excuse for people to democratize data access. I wrote about this aspect here: https://x.com/iosifache/status/1941049600162574676?s=46
And BTW, that’s a good idea! The functionality should probably also be exposed via CLI.
Claude Desktop also has a built-in file reader [2], so you can ask it to read the file and process the content (e.g., generate a summary or even a meta-summary [3]).
[1] https://github.com/sylphxltd/pdf-reader-mcp [2] https://github.com/modelcontextprotocol/servers/tree/main/sr... [3] https://x.com/iosifache/status/1942247320302547175
There's no way to protect against this. Anna's Archive doesn't include licence information in their data fields. It would be helpful to integrate with another data source that could warn MCP server users when they're attempting potentially risky actions. Please let me know if you have ideas on how to achieve this.
On a related note, please see this reply:
The comparison might be loose, but the problem is similar to releasing a browser. Do you prevent users from accessing websites you think are malicious or illegal? Or do you delegate that responsibility?
I was hesitant about releasing the MCP server as open source software, but I hope (1) it proves useful for others and (2) people understand that the authors of the books they're reading need money to eat, live, and support their families.
Are you referring to the JSON index (https://annas-archive.org/faq#api)?
Regarding the second last, there’s a chance that only one person will remember who the other person is. If you have a super intense networking night, things might get a bit hazy.
But these are fair thoughts, so cheers for the feedback!
Regardless of whether you’re a developer looking to design secure software or simply undertaking application security tasks, this news might interest you:
I’ve compiled an awesome list of tools that can be easily installed on Ubuntu and used to safeguard your codebases.
As the repository is open source, please feel free to recommend any application security tools that you find useful. PRs are welcome!
Based on a custom, purposefully vulnerable Python and C codebase, I proposed tasks using a variety of techniques and tools:
- Threat modelling with OWASP Threat Dragon;
- Secret scanning with Gitleaks;
- Dependency scanning with OSV-Scanner;
- Linting with Bandit and flawfinder;
- Code querying with Semgrep;
- Fuzzing with AFL++; and
- Symbolic execution with KLEE.
The workshop consists of an online wiki and a GitHub repository with source code and pre-built Docker images.
It is meant to be solved at home without the live assistance of a workshop host. Just follow the next steps:
1. Review the concepts of SDLC and software security.
2. Understand and set up the analysis infrastructure.
3. Understand the vulnerable application that will be analysed: its functionality, architecture, and vulnerabilities.
4. For each analysis technique, solve the proposed tasks. If encountering blockers, the proposed solutions can be used.
5. Review what other analysis techniques exist and how all techniques can be automated.
6. Review the security checklist and think about how the techniques and tools can be embedded in the development process of participant's projects.
Please let me know what you think about it!
If you need support or have a question or proposal, reach out to me, or just create an issue in the GitHub repository.