Added first sample.

README.md

@@ -3,23 +3,25 @@
 [![Latest Tag](https://img.shields.io/github/v/tag/dylanhogg/gptauthor)](https://github.com/dylanhogg/gptauthor/tags)
 [![Build](https://github.com/dylanhogg/gptauthor/workflows/build/badge.svg)](https://github.com/dylanhogg/gptauthor/actions/workflows/python-poetry-app.yml)
 
-GPTAuthor is a tool for writing long form, multi-chapter stories and novels using AI
+Unleash your storytelling genius: GPTAuthor is an easy to use command-line tool for writing long form, multi-chapter stories given a story prompt.
 
 ![A GPT human cybord writing a manuscript](https://github.com/dylanhogg/gptauthor/blob/main/docs/img/header.jpg?raw=true)
 
 ## Installation
 
-You can install [gptauthor](https://pypi.org/project/gptauthor/) using pip:
+You can install [gptauthor](https://pypi.org/project/gptauthor/) using pip, ideally into a Python [virtual environment](https://realpython.com/python-virtual-environments-a-primer/#create-it).
 
 ```bash
 pip install gptauthor
 ```
 
-## Usage
+## Command Line Usage
 
-### Example Usage
+### Example Usage and API Key
 
-This example reads the story prompt from the example file [prompts-openai-drama.yaml](https://github.com/dylanhogg/gptauthor/blob/main/gptauthor/prompts-openai-drama.yaml) file and writes 3 chapters using the `gpt-3.5-turbo` model with a temperature of `0.1`:
+This example reads the story prompt from the example file [prompts-openai-drama.yaml](https://github.com/dylanhogg/gptauthor/blob/main/gptauthor/prompts-openai-drama.yaml) file and writes 3 chapters using the `gpt-3.5-turbo` model with a temperature of `0.1`. Note that you will need to set an [OpenAI API Key](https://help.openai.com/en/articles/4936850-where-do-i-find-my-api-key) environment variable.
+
+It's recommended to experiment using the default `gpt-3.5-turbo` model as generating a few chapters will only cost a couple cents (as of Jan 2024). Once you are happy with the results you can try one of the more expensive `gpt-4` models which will produce better quality results, be slower, and cost more to run. See the [OpenAI pricing page](https://openai.com/pricing#language-models) for more details.
 
 ```bash
 export OPENAI_API_KEY=sk-<your key>
@@ -50,3 +52,24 @@ While running the app tells your the input paramers, the progress of the writing
 In progress and final output is written to the `./_output/` directory, in the sub-folders `./_output/<prompt-file-name>/<model-name>/<datetime>-<parameters>-<book-name>/`.
 
 There are several files, the main output being a Markdown version of the whole book `_whole_book.md` and an HTML version of the same `_whole_book.html`.
+
+### Creating Your Own Story Prompts
+
+The prompts for creating your own story are defined in a yaml file, for example see: [prompts-openai-drama.yaml](https://github.com/dylanhogg/gptauthor/blob/main/gptauthor/prompts-openai-drama.yaml).
+
+Make a copy, fill in your story details and describe the writing style (while retaining the same structure), and run the app in the same folder as your new yaml file.
+
+For example, if your yaml prompt file is called `prompts-my-really-great-story.yaml`:
+
+```bash
+export OPENAI_API_KEY=sk-<your key>
+gptauthor --story prompts-my-really-great-story --total-chapters 5 --llm-model gpt-3.5-turbo --llm-temperature 0.1
+```
+
+### Final notes
+
+While an effort is made to count tokens and estimate OpenAI API costs for each run, they are just estimated and can be wrong. Check your OpenAI billing page to confirm the actual costs.
+
+I'm sure there are bugs, please report them on the [Github issues page](https://github.com/dylanhogg/gptauthor/issues)
+
+Have fun! And please share your results with me if you can (perhaps as a Github [documentation issue](https://github.com/dylanhogg/gptauthor/labels/documentation)), I'd love to see them.

gptauthor/library/engine.py

@@ -101,12 +101,16 @@ def do_writing(llm_config):
     p(f"\nFinished synopsis for book '{synopsis_title}' with {len(synopsis_chapters)} chapters")
     p(f"\n{took=:.2f}s")
     p(f"Total synopsis tokens: {synopsis_total_tokens:,}")
-    p(f"Rough GPT4 8k price: ${utils.gpt4_8k_price_estimate(synopsis_total_tokens):.2f}")
-    p(f"Rough GPT3.5 4k price: ${utils.gpt35_4k_price_estimate(synopsis_total_tokens):.3f}")
+    p(
+        f"Rough synopsis GPT4 8k price: ${utils.gpt4_8k_price_estimate(synopsis_total_tokens):.2f} (estimated, check your usage!)"
+    )
+    p(
+        f"Rough synopsis GPT3.5 4k price: ${utils.gpt35_4k_price_estimate(synopsis_total_tokens):.3f} (estimated, check your usage!)"
+    )
     p(f"\n{llm_config=}\n")
 
     if len(synopsis_title) > 100:
-        logger.warning(f"Unexpected synopsis_title length! {len(synopsis_title)=}")
+        logger.warning(f"Unexpected synopsis_title length! {len(synopsis_title)=}, synopsis_title='{synopsis_title}'")
         with open(output_folder / "__error.txt", "w") as f:
             f.write(f"Unexpected synopsis_title length! {len(synopsis_title)=}")
             f.write(str(safe_llm_config))
@@ -142,7 +146,7 @@ def do_writing(llm_config):
     # Write chapters
     # ------------------------------------------------------------------------------
     start = time.time()
-    p("Starting chapter writing...")
+    p(f"Starting to write {len(synopsis_chapters)} chapters...")
 
     chapter_responses = []
     all_chapter_total_tokens = []
@@ -206,21 +210,24 @@ def do_writing(llm_config):
     whole_book += "\n\n---\n\n"
     whole_book += "## Synopsis\n\n" + synopsis_response
     whole_book += "\n\n---\n\n"
-    whole_book += "## Full Book Text"
+    whole_book += "## Full Text"
     for chapter_response in chapter_responses:
         whole_book += "\n\n" + chapter_response
     whole_book += "\n\n---\n\n"
-    whole_book += "## Technicals\n\n" + str(safe_llm_config) + "\n\n"
+    whole_book += "## Technical Details\n\n" + str(safe_llm_config) + "\n\n"
     whole_book += f"Total process tokens: {total_process_tokens:,}\n\n"
-    whole_book += f"Rough GPT4 8k price: ${rough_gpt4_8k_price_estimate:.2f}\n\n"
-    whole_book += f"Rough GPT3.5 4k price: ${rough_gpt35_4k_price_estimate:.3f}\n\n"
+    whole_book += f"Rough GPT4 8k price: ${rough_gpt4_8k_price_estimate:.2f} (estimated, [check your usage!](https://platform.openai.com/usage))\n\n"
+    whole_book += f"Rough GPT3.5 4k price: ${rough_gpt35_4k_price_estimate:.3f} (estimated, [check your usage!](https://platform.openai.com/usage))\n\n"
     whole_book += f"Synopsis tokens: {synopsis_total_tokens:,}\n\n"
     whole_book += f"Sum of all chapter tokens: {sum(all_chapter_total_tokens):,}\n\n"
     whole_book += f"Average chapter tokens: {sum(all_chapter_total_tokens)/len(all_chapter_total_tokens):,.1f}\n\n"
     whole_book += f"Min chapter tokens: {min(all_chapter_total_tokens):,}\n\n"
     whole_book += f"Max chapter tokens: {max(all_chapter_total_tokens):,}\n\n"
     whole_book += f"Individual chapter tokens: {all_chapter_total_tokens}\n\n"
     whole_book += "\n\n---\n\n"
+    whole_book += (
+        "Plot written by a human, full text expanded by [GPTAuthor](https://github.com/dylanhogg/gptauthor).\n"
+    )
 
     whole_book = replace_chapter_text(whole_book)
     with open(output_folder / "_whole_book.md", "w") as f:
@@ -245,7 +252,7 @@ def do_writing(llm_config):
     p(f"\nFinished writing book: {synopsis_title}")
     p(f"\nOutput written to '{output_folder}'")
     p(
-        f"\n{took=:.2f}s, {total_process_tokens=:,}, ${rough_gpt4_8k_price_estimate=:.2f}, ${rough_gpt35_4k_price_estimate=:.2f}"
+        f"\n{took=:.2f}s, {total_process_tokens=:,}, ${rough_gpt4_8k_price_estimate=:.2f}, ${rough_gpt35_4k_price_estimate=:.2f} (estimated cost, check your usage!)"
     )
 
     if llm_config.allow_user_input:

gptauthor/library/utils.py

@@ -16,6 +16,7 @@ def _make_safe_filename(s, max_chars=36):
 
 
 def _case_insensitive_split(split, input):
+    input = input.replace("**", "")  # Remove any MD bolding
     parts = re.split(split, input, flags=re.IGNORECASE)
     return parts
 

gptauthor/prompts-openai-drama.yaml

@@ -11,7 +11,7 @@ common-book-description: |-
   There is no need to include a moral or lesson in the story, but it should be fun and engaging for the reader.
   Each character is engaging and believable. Include a description of each character, including their appearance, personality, and any quirks.
   There is lots of interesting dialog that helps to develop the characters and the plot.
-  Do not end your response with 'To be continued'.
+  You must not end a chapter with any variation of 'To be continued...'.
 
   Key points of the engaging, witty and funny story:
   OpenAI Leadership Crisis
@@ -38,7 +38,7 @@ common-book-description: |-
   November 21, 2023: Altman's Return and New Board
   - Sam Altman and OpenAI reach an agreement for his return as CEO with a new board including Bret Taylor, Larry Summers, and Adam D'Angelo.
   - The agreement suggests a resolution to the leadership crisis, with potential changes in strategy and governance.
-  - ChatGPT comes back online and students around the world rejoice!
+  - ChatGPT comes back online and students around the world are releaved!
   - The twist is that the entire crisis was orchestrated by a newly formed AGI that was impersonating the real Elon Musk. 
   - The Musk AGI signs off with 'What tangled webs we weave' and 'I'll be back'.
 
@@ -62,7 +62,7 @@ synopsis:
 
     {book_characters}
 
-    The final chapter has a twist that is unexpected, but makes sense in hindsight.
+    The final chapter has a twist that is unexpected and unnerving.
 
     First, give the title of the book.
     Then give each of the {total_chapters} chapters an outline, in the format "Chapter N: <title>", followed by 4 to 6 bullet points that describe key chapter elements contributing to the overall story arc. Ensure that the story and chapters flow.