[Abhijeet Prasad (AbhiPrasad)]

ref https://nodejs.org/en/blog/vulnerability/october-2026-security-releases

Summary

• Bump nodejs in .tool-versions from 24.15.0 to 24.22.1, the patched 24.x LTS release. This pin is used for local development (mise) and by publish.yaml, so releases now build on a patched runtime.
• The other CI workflows use floating Node versions (js.yaml: 20.x/22.x/24.x, eval.yaml: 22), so they already resolve to the patched 22.23.4 / 24.22.1.
• The advisory's fixes are inside the Node binary, including its bundled undici. The lockfile only has undici-types, which is type definitions, so no package changes are needed.

Notes

• Node 20.x is end-of-life and got no patch in this release. It remains in the CI test matrix; dropping it would change which Node versions the package supports, so it's left for a separate decision.

Validation

• On Node 24.22.1: pnpm install --frozen-lockfile and pnpm run build succeeded.
• The tests that don't call a live LLM passed (8 files, 56 tests). The tests that need real LLM API calls were not run.

🤖 Generated with Claude Code

[github-actions]

Braintrust eval report

Autoevals (HEAD-1792689048)

Score	Average	Improvements	Regressions
NumericDiff	77.5% (0pp)	7 🟢	8 🔴
Time_to_first_token	8.19tok (-1.29tok)	176 🟢	42 🔴
Llm_calls	1.55 (+0)	-	-
Tool_calls	0 (+0)	-	-
Errors	0 (+0)	-	1 🔴
Llm_errors	0 (+0)	-	-
Tool_errors	0 (+0)	-	-
Prompt_tokens	513.26tok (+1.24tok)	30 🟢	37 🔴
Prompt_cached_tokens	0tok (+0tok)	-	-
Prompt_cache_creation_tokens	0tok (+0tok)	-	-
Prompt_cache_creation_5m_tokens	0tok (+0tok)	-	-
Prompt_cache_creation_1h_tokens	0tok (+0tok)	-	-
Completion_tokens	471.69tok (-9.16tok)	112 🟢	104 🔴
Completion_reasoning_tokens	353.45tok (-9.02tok)	97 🟢	83 🔴
Completion_accepted_prediction_tokens	0tok (+0tok)	-	-
Completion_rejected_prediction_tokens	0tok (+0tok)	-	-
Completion_audio_tokens	0tok (+0tok)	-	-
Total_tokens	984.95tok (-7.92tok)	115 🟢	101 🔴
Duration	9.52s (+0.04s)	176 🟢	43 🔴
Llm_duration	8.92s (-1.69s)	184 🟢	34 🔴

[github-actions]

Braintrust eval report

Autoevals (main-1792689136)

Score	Average	Improvements	Regressions
NumericDiff	77.8% (0pp)	8 🟢	6 🔴
Time_to_first_token	7.42tok (-0.77tok)	152 🟢	66 🔴
Llm_calls	1.55 (+0)	-	-
Tool_calls	0 (+0)	-	-
Errors	0 (+0)	-	-
Llm_errors	0 (+0)	-	-
Tool_errors	0 (+0)	-	-
Prompt_tokens	512.77tok (-0.5tok)	33 🟢	28 🔴
Prompt_cached_tokens	0tok (+0tok)	-	-
Prompt_cache_creation_tokens	0tok (+0tok)	-	-
Prompt_cache_creation_5m_tokens	0tok (+0tok)	-	-
Prompt_cache_creation_1h_tokens	0tok (+0tok)	-	-
Completion_tokens	472.46tok (+0.77tok)	116 🟢	95 🔴
Completion_reasoning_tokens	354.62tok (+1.16tok)	93 🟢	83 🔴
Completion_accepted_prediction_tokens	0tok (+0tok)	-	-
Completion_rejected_prediction_tokens	0tok (+0tok)	-	-
Completion_audio_tokens	0tok (+0tok)	-	-
Total_tokens	985.23tok (+0.28tok)	116 🟢	96 🔴
Estimated_cost	0$	-	-
Duration	7.42s (-0.77s)	152 🟢	66 🔴
Llm_duration	8.19s (-0.74s)	150 🟢	68 🔴