ranxianglei/billion-context

C3

A context-compression plugin for month-long single sessions (billions of tokens) ,small context windows (a 100K context is enough) and token savings (5x fewer tokens) .上下文压缩插件,月级长会话会话, 兼顾小窗口(100k上下文足矣

★ 5.5k · ranxianglei/billion-context source on GitHub · this plugin in the registry

ranxianglei/billion-context is a DeepSeek Harness plugin rated C3 — powerful capability combined with sensitive behavior. It patches the dsh runtime, executes system commands, decodes base64 payloads, starts a network server.

Installable plugin — declares a dsh.bundle manifest

What it can do

CapabilityFlagEvidence
patches the dsh runtimeruntime_patch./dsh.bundle.patch.yml
executes system commandsexec×99 in authored code, e.g. scripts/check-dist-thin-install.mjs:14, scripts/check-dist-thin-install.mjs:14
decodes base64 payloadsbase64_decode×9 in authored code, e.g. scripts/blog-discussions.mjs:100, src/apig-resign.ts:195
starts a network servernet_server×16 in authored code, e.g. scripts/release-gate-local.mjs:105, scripts/release-gate-local.mjs:106
reads credential-class env varstoken_envACP_COUNT_TOKENS_PASSTHROUGH, ACP_PASSTHROUGH, BILI_DSH_AUTHORED_COMPACTION, BILI_ENCRYPTION_KEY, BILI_IMAGE_TOKEN_CAP, BILI_LAUNCH_TOKEN

Tool registrations

0 confirmed registration point(s); analysis is incomplete.

1203 source file(s) omitted by size/count/encoding limits; tool count is incomplete

Services it injects

agents commands locale slots tools

Hooks it attaches

internal/config session/created session/event

Outbound domains

api.anthropic.com api.openai.com api.github.com chatgpt.com models.dev tencent.sso.codebuddy.cn registry.npmmirror.com open.bigmodel.cn billion-context.inline generativelanguage.googleapis.com

Environment variables it reads

BILLION_CONTEXT_PROXY NODE_TEST_CONTEXT BILI_MCP_PROXY BILI_ENCRYPTION_KEY BILI_PERSIST_ZSTD GITHUB_TOKEN https_proxy HTTPS_PROXY http_proxy HTTP_PROXY

How to read this

Levels measure capability surface and transparency, not maliciousness. A C3 plugin can be entirely legitimate — a desktop shell genuinely needs subprocesses. The point is that you can see this before installing. See the levels explained and how dsh plugins work.

Findings come from static analysis of shipped code; nothing is executed. Think a flag is wrong? Open an issue — every flag cites the file and line it came from.