Docs/limit context window 1304

[Sdsai0311]

changed the code

---

Important

This PR updates documentation with new context window strategies and improves installation and usage instructions, while modifying pre-commit configuration for enhanced code quality checks.

• Documentation:
  • Adds context_window.md to explain context window management strategies.
  • Updates README.md with installation, setup, and usage instructions.
  • Corrects links and instructions in docs/examples/open_llms/README.md, docs/index.rst, docs/introduction.md, docs/open_models.md, and docs/quickstart.rst.
• Pre-commit Configuration:
  • Changes default_stages from [commit] to [pre-commit] in .pre-commit-config.yaml.

^{This description was created by for 218b041. You can customize this summary. It will automatically update as commits are pushed.}

[ellipsis-dev]

Caution

Changes requested ❌

Reviewed everything up to 218b041 in 1 minute and 45 seconds. Click for details.

• Reviewed 433 lines of code in 8 files
• Skipped 0 files when reviewing.
• Skipped posting 7 draft comments. View those below.
• Modify your settings and rules to customize what types of comments Ellipsis leaves. And don't forget to react with 👍 or 👎 to teach Ellipsis.

1. .pre-commit-config.yaml:4

• Draft comment:
  Ensure that changing 'default_stages' from [commit] to [pre-commit] is intentional. This change affects when hooks are triggered, so verify that it matches your workflow requirements.
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 10% vs. threshold = 50% This is exactly the type of comment we should remove. It's asking the author to verify their intention and double-check things, which violates our rules. The change is in a configuration file where changes are typically intentional. There's no evidence of an actual problem. Maybe this is a critical configuration change that could break the pre-commit workflow for other developers? Even if it is an important change, asking for verification doesn't help - we should trust that the author made this change deliberately, and if there are issues, they will be discovered during normal usage. Delete this comment as it violates our rule about not asking authors to verify their intentions or double-check things.

2. README.md:10

• Draft comment:
  Typo: 'genereation' should be corrected to 'generation'.
• Reason this comment was not posted:
  Comment was not on a location in the diff, so it can't be submitted as a review comment.

3. docs/introduction.md:17

• Draft comment:
  The video tag's src attribute appears to be broken across lines. Ensure the URL is continuous (with no newline) so that the video loads correctly.
• Reason this comment was not posted:
  Comment was not on a location in the diff, so it can't be submitted as a review comment.

4. docs/context_window.md:37

• Draft comment:
  In the pseudocode, 'payload.prepend(msg)' is used. Consider clarifying or using a more standard term (like 'insert at beginning') since 'prepend' may be ambiguous in some languages.
• Reason this comment was not posted:
  Decided after close inspection that this draft comment was likely wrong and/or not actionable: usefulness confidence = 10% vs. threshold = 50% This is pseudocode in documentation, not actual implementation code. The term 'prepend' is widely understood in programming contexts and is actually clearer for explaining the concept than 'insert(0)'. The comment seems overly pedantic given this is conceptual documentation. Maybe some developers are not familiar with the term 'prepend' and would benefit from more standard terminology. The code is clearly labeled as conceptual pseudocode, and 'prepend' is a common term in programming that clearly communicates the intent. Making it more "standard" would actually make it less readable. Delete the comment as it suggests an unnecessary change to conceptual pseudocode that would make the documentation less clear.

5. docs/examples/open_llms/README.md:56

• Draft comment:
  The relative link has been updated to '../../open_models.md'. Please verify that this relative path resolves correctly in all intended documentation contexts.
• Reason this comment was not posted:
  Comment did not seem useful. Confidence is useful = 0% <= threshold 50% This comment is asking the PR author to verify the correctness of a relative path, which falls under the category of asking the author to double-check something. This violates the rule against asking the author to confirm or ensure something.

6. docs/quickstart.rst:32

• Draft comment:
  Review the relative links (e.g., 'open_models.md' and 'windows_readme_link.rst') to ensure they are correct and that the target files exist in the built documentation.
• Reason this comment was not posted:
  Comment did not seem useful. Confidence is useful = 0% <= threshold 50% This comment is asking the PR author to ensure that the relative links are correct and that the target files exist in the built documentation. This falls under the category of asking the author to double-check things, which is against the rules.

7. README.md:14

• Draft comment:
  Typographical inconsistency: The sentence "If you are looking for a well maintained hackable CLI for – check out aider." includes the word "for", whereas the later instance omits it. Please verify the intended phrasing for consistency.
• Reason this comment was not posted:
  Comment did not seem useful. Confidence is useful = 20% <= threshold 50% The comment is pointing out a typographical inconsistency in the phrasing of a sentence. However, it is asking the author to verify the intended phrasing, which is against the rules. The comment does not provide a specific suggestion or ask for a specific change, making it more of a request for confirmation rather than a constructive suggestion.

Workflow ID: wflow_dhchmRN4u6c5cbeV

^{You can customize by changing your verbosity settings, reacting with 👍 or 👎, replying to comments, or adding code review rules.}