AprilWizard/dsh-multi-cot
C3Multi-CoT plugin for DeepSeek Harness: multi-sampled test-time compute, internal voting, and a plan/execute/review workflow
★ 1+ · AprilWizard/dsh-multi-cot source on GitHub · this plugin in the registry
AprilWizard/dsh-multi-cot is a DeepSeek Harness plugin rated C3 — powerful capability combined with sensitive behavior. It can rewrite the system prompt, reads credential-class env vars, ships no manifest.
What it can do
| Capability | Flag | Evidence |
|---|---|---|
| can rewrite the system prompt | prompt_surface | systemPrompt |
| reads credential-class env vars | token_env | OPENCODE_API_KEY, OPENCODE_GO_API_KEY |
| ships no manifest | no_manifest | uses 4 services, 2 tool regs, no manifest |
Services it injects
invariants llm sessions systemPrompt
Hooks it attaches
agent/pre-step agent/status agent/turn-stopping internal/dispatch llm/stream session/created
Outbound domains
opencode.ai
Environment variables it reads
OPENCODE_GO_BASE_URL OPENCODE_GO_API_KEY OPENCODE_API_KEY OPENCODE_GO_MODEL
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.