[PR #97] [CLOSED] Update dependencies to resolve max out tokens issue #116

New issue

Closed

opened 2026-02-27 08:09:16 +03:00 by kerem · 0 comments

kerem commented

2026-02-27 08:09:16 +03:00

Owner

📋 Pull Request Information

Original PR: https://github.com/KeygraphHQ/shannon/pull/97
Author: @nsimmons
Created: 2/8/2026
Status: ❌ Closed

Base: main ← Head: upgrade_claude_sdk

📝 Commits (1)

b63f380 Update dependencies to resolve max out tokens issue

📊 Changes

5 files changed (+191 additions, -41 deletions)

View changed files

📝 mcp-server/package-lock.json (+87 -9)
📝 mcp-server/package.json (+2 -2)
📝 package-lock.json (+94 -26)
📝 package.json (+3 -3)
📝 src/ai/claude-executor.ts (+5 -1)

📄 Description

Summary

Fix for https://github.com/KeygraphHQ/shannon/issues/65, and alternate solution from https://github.com/KeygraphHQ/shannon/pull/75

@anthropic-ai/claude-agent-sdk 0.1.0 gem has a hard limit of 32k output tokens. This limit means that setting CLAUDE_CODE_MAX_OUTPUT_TOKENS to > 32k has no impact. Upgrading to the latest version solves this issue as that limitation has been removed.

Changes

Upgraded @anthropic-ai/claude-agent-sdk 0.1.0 and dependent zod gems to latest versions.
Modified src/ai/claude-executor.ts to pass CLAUDE_CODE_MAX_OUTPUT_TOKENS in the way newest version of claude agent SDK expects.

Test Plan

Set CLAUDE_CODE_MAX_OUTPUT_TOKENS to 64k
Run ./shannon start on a repo that generates a pre-recon output that is larger than 32k
Verify deliverable is created without any output token errors

I understand you're not accepting external PRs. I have found this tool very useful and wanted to do what I can to suggest improvements even if they are not accepted. If it helps you resolve an issue faster then I have done my part to contribute back.

_{🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.}

## 📋 Pull Request Information **Original PR:** https://github.com/KeygraphHQ/shannon/pull/97 **Author:** [@nsimmons](https://github.com/nsimmons) **Created:** 2/8/2026 **Status:** ❌ Closed **Base:** `main` ← **Head:** `upgrade_claude_sdk` --- ### 📝 Commits (1) - [`b63f380`](https://github.com/KeygraphHQ/shannon/commit/b63f3803c176271b08ba1b4b696105e88ad18548) Update dependencies to resolve max out tokens issue ### 📊 Changes **5 files changed** (+191 additions, -41 deletions) <details> <summary>View changed files</summary> 📝 `mcp-server/package-lock.json` (+87 -9) 📝 `mcp-server/package.json` (+2 -2) 📝 `package-lock.json` (+94 -26) 📝 `package.json` (+3 -3) 📝 `src/ai/claude-executor.ts` (+5 -1) </details> ### 📄 Description # Summary Fix for https://github.com/KeygraphHQ/shannon/issues/65, and alternate solution from https://github.com/KeygraphHQ/shannon/pull/75 `@anthropic-ai/claude-agent-sdk 0.1.0` gem has a hard limit of 32k output tokens. This limit means that setting `CLAUDE_CODE_MAX_OUTPUT_TOKENS` to > 32k has no impact. Upgrading to the latest version solves this issue as that limitation has been removed. # Changes - Upgraded `@anthropic-ai/claude-agent-sdk 0.1.0` and dependent `zod` gems to latest versions. - Modified `src/ai/claude-executor.ts` to pass `CLAUDE_CODE_MAX_OUTPUT_TOKENS` in the way newest version of claude agent SDK expects. # Test Plan - [ ] Set `CLAUDE_CODE_MAX_OUTPUT_TOKENS` to 64k - [ ] Run `./shannon start` on a repo that generates a pre-recon output that is larger than 32k - [ ] Verify deliverable is created without any output token errors I understand you're not accepting external PRs. I have found this tool very useful and wanted to do what I can to suggest improvements even if they are not accepted. If it helps you resolve an issue faster then I have done my part to contribute back. --- <sub>🔄 This issue represents a GitHub Pull Request. It cannot be merged through Gitea due to API limitations.</sub>