Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
Python agentaianthropicclaude-codecompressioncontext-engineeringcontext-windowcursorfastapilangchainllmmcpopenaiprompt-engineeringproxypythonragtoken-optimizationtokenstypescript
ci: pin least-privilege GITHUB_TOKEN permissions (contents: read)
Tejas Chopra6f0dc32680bd
diff --git a/.github/workflows/ci-fast.yml b/.github/workflows/ci-fast.ymlindex a5751fdf8..cbb5543db 100644--- a/.github/workflows/ci-fast.yml+++ b/.github/workflows/ci-fast.yml@@ -17,6 +17,13 @@ on:branches: [main]workflow_dispatch:+# Least-privilege GITHUB_TOKEN: every job here only reads the repo (checkout,+# paths-filter, build, test). No job writes contents/PRs/releases, so read-only+# is sufficient and satisfies the CodeQL "workflow does not contain permissions"+# advisory.+permissions:+ contents: read+concurrency:group: ci-fast-${{ github.ref }}cancel-in-progress: true