Destiny: Blending Classical Astrology with Claude Code's AI for Fortune Telling

The emergence of large language models (LLMs) has opened new avenues for integrating AI into diverse applications, even those rooted in ancient traditions. One such intriguing project is "Destiny," a Claude Code plugin designed to provide fortune readings based on classical East Asian astrology. This tool stands out by attempting to bridge the deterministic nature of traditional astrological calculations with the interpretive capabilities of an LLM, specifically Claude.

Destiny offers a fascinating case study in how AI can be leveraged to process structured data and generate nuanced, context-aware prose. It prompts important discussions about the role of AI in sensitive domains like personal advice, the challenge of avoiding hallucination, and the practical considerations of deploying such tools in development environments.

How Destiny Works: A Two-Layered Approach

Destiny operates within the Claude Code interface, allowing users to invoke a fortune reading by simply typing /destiny. Upon first use, users provide their birthday, which is then stored for future readings. The core innovation lies in its two-layered architecture, designed to maintain fidelity to traditional astrological systems while leveraging AI for interpretation:

Layer 1: Deterministic Astrological Calculation

The foundational layer of Destiny is a Python script responsible for all numerical calculations. This script computes several key astrological elements based on the user's birth date and the current day:

• Eight-character birth chart: A fundamental component of East Asian astrology.
• Today's day pillar: Represents the astrological influences of the current day.
• Hexagram for the moment: Derived from I Ching principles.
• Five-element relationships: Interactions between the five classical elements (wood, fire, earth, metal, water).

The developer emphasizes that this layer is entirely deterministic: "Same person + same day = identical output." This ensures that the raw astrological data is consistent and verifiable against traditional calendar sources, preventing any AI-driven fabrication at the data generation stage.

Layer 2: Claude's Interpretive Prose

Once the deterministic data is generated, Claude, the LLM, takes over. Its role is to translate these numerical and symbolic inputs into human-readable prose. This includes generating:

• Today's stars: Astrological influences specific to the day.
• Character sketch: Insights into the user's personality.
• Life arc: General trends and potential paths.
• Advice: Guidance based on the astrological reading.

The crucial distinction here, as highlighted by the developer, is that Claude applies "centuries-old reading conventions to that fixed data." This approach aims to prevent "LLM-hallucinated horoscope" by grounding the AI's output in established interpretive frameworks, rather than allowing it to invent details freely.

Community Insights and Discussions

The launch of Destiny on Hacker News sparked a variety of reactions and discussions, touching upon its utility, technical implementation, and broader implications.

Audience and Use Case Relevance

One common thread of discussion revolved around the target audience and the appropriateness of a fortune-telling plugin within a coding environment.

"interesting, but I really wonder how big the audience is of people wanting to know their fortune while claude coding within the claude code interface" — @treexs

While some found it fun, others questioned the practical demand for such a tool in a developer's workflow.

Installation and Usability

Early feedback pointed to the need for clearer installation instructions, particularly for users unfamiliar with the Claude plugin marketplace.

"Could you flesh out the installation instructions a little more? I haven't used the plugin marketplace yet, and don't have a /plugin command. Claude itself is not able to tell me how to install it." — @awakeasleep

This highlights a common challenge for new plugin developers: ensuring a smooth onboarding experience for users across varying levels of technical familiarity with the platform.

Accuracy, Hallucination, and Trust

The claim of avoiding "LLM-hallucinated horoscopes" naturally led to skepticism and questions about the validity of AI-generated interpretations versus traditional methods.

"I am resisting the urge to be very snarky. But how do you know that Claude's hallucinated horoscopes are any less accurate than traditional techniques? Have you A/B tested them?" — @sudosteph

This comment underscores the ongoing debate about the perceived accuracy of LLM outputs, especially when dealing with subjective or interpretive content. Another user aptly summarized the project's core dynamic:

"So the astrology is deterministic, and the interpretation is probabilistic. Which is interesting, because that’s exactly where most of the perceived “accuracy” tends to come from." — @tokyoproductj

This observation points to the inherent challenge and potential for bias in any interpretive system, whether human or AI.

Ethical and Legal Considerations

The nature of providing "advice" through an LLM raised concerns about potential negative outcomes, prompting suggestions for legal disclaimers.

"Since you seem to be mentioning 'destiny', 'life arc' and 'advice' -it may be wise to put in some kind of a legal disclaimer for your own sake -- the use of LLM toys for advice about life and similar have led to some really bad outcomes, even suicides as we know now." — @hansmayer

This highlights the critical need for developers of AI tools offering personal guidance to consider the ethical implications and potential liabilities.

Security and Resource Usage

Other comments touched upon broader technical concerns, including potential "supply chain risk" from integrating third-party plugins and the consumption of LLM tokens for non-essential tasks.

"Looks like a supply chain risk to me." — @pdp

"I thought anthropic is tightening usage but it seems people have nothing to do with their tokens …" — @chakintosh

Potential Enhancements and Future Directions

Several users offered constructive suggestions for extending Destiny's functionality:

• Automated Invocation: Integrating "Claude code hooks" to trigger fortune readings at specific agent lifecycle events, such as SessionStart, was proposed as an alternative to manual invocation. (@tomaytotomato)
• Multi-User Profiles: A "marriage palace built using two user profiles" was suggested as a feature to compare astrological charts between individuals. (@babblingfish)
• LLM Comparability: Curiosity was expressed regarding how the output might differ if generated by other LLMs, hinting at the potential for comparative evaluations. (@gorlokthedev)

Conclusion

Destiny represents a novel application of LLMs, demonstrating how AI can be integrated with structured, traditional systems to generate personalized content. By separating deterministic data calculation from probabilistic interpretation, it attempts to navigate the complex landscape of AI accuracy and hallucination. The community's response underscores both the playful potential of such tools and the serious considerations around their ethical implications, usability, and technical robustness. As LLM capabilities continue to evolve, projects like Destiny offer valuable insights into the diverse and sometimes unexpected ways AI can intersect with human culture and tradition.