DeepSeek Harness in VSCode is a skill AI agent for To get instant AI-generated code recommendations and completions while writing code. Pricing: not stated on the product page. As of 2026-10-27, KanonAgent records 2 upvotes. First indexed by KanonAgent on 2026-08-21.
DeepSeek Harness in VSCode brings DeepSeek's models directly into the developer's editor to provide real-time code suggestions and completions.
2 upvotes
Tracked by Kanon since Aug 21, 2026
Status basis: npm_downloads_7d gained 33680 in the last 3d after a flat 3d baseline (no prior movement to compare a ratio against)
📈 Timelinewhat changed, and when
- Accelerating npm_downloads_7d gained 33680 in the last 3d after a flat 3d baseline (no prior movement to compare a ratio against)
- Cooling npm_downloads_7d momentum fell to 0 per 3d from 54932 (<= 40% of the prior window)
- Cooling npm_downloads_7d momentum fell to -131260 per 3d from 17760 (<= 40% of the prior window)
- Cooling npm_downloads_7d momentum fell to -131260 per 3d from 20954 (<= 40% of the prior window)
- Cooling npm_downloads_7d momentum fell to 0 per 3d from 47269 (<= 40% of the prior window)
- Cooling npm_downloads_7d momentum fell to 17760 per 3d from 59824 (<= 40% of the prior window)
- Cooling npm_downloads_7d momentum fell to 20954 per 3d from 69556 (<= 40% of the prior window)
- Cooling npm_downloads_7d momentum fell to 20954 per 3d from 69556 (<= 40% of the prior window)
- Cooling npm_downloads_7d momentum fell to -57858 per 3d from 31430 (<= 40% of the prior window)
- Cooling npm_downloads_7d momentum fell to -52817 per 3d from 38296 (<= 40% of the prior window)
- Cooling npm_downloads_7d momentum fell to -30907 per 3d from 9588 (<= 40% of the prior window)
- Accelerating npm_downloads_7d gained 31430 in the last 3d vs 17201 in the 3d before (ratio 1.8x, threshold 1.5x)
- Cooling npm_downloads_7d momentum fell to -30907 per 3d from 9588 (<= 40% of the prior window)
- Cooling npm_downloads_7d momentum fell to 38296 per 3d from 149923 (<= 40% of the prior window)
- Accelerating npm_downloads_7d gained 31430 in the last 3d vs 17201 in the 3d before (ratio 1.8x, threshold 1.5x)
- Cooling npm_downloads_7d momentum fell to 9588 per 3d from 156721 (<= 40% of the prior window)
- Cooling npm_downloads_7d momentum fell to 17201 per 3d from 121736 (<= 40% of the prior window)
- Accelerating npm_downloads_7d gained 149923 in the last 3d after a flat 3d baseline (no prior movement to compare a ratio against)
- Cooling npm_downloads_7d momentum fell to 17201 per 3d from 121736 (<= 40% of the prior window)
- Accelerating npm_downloads_7d gained 156721 in the last 3d vs 12619 in the 3d before (ratio 12.4x, threshold 1.5x)
- Accelerating npm_downloads_7d gained 121736 in the last 3d vs 7673 in the 3d before (ratio 15.9x, threshold 1.5x)
- Cooling npm_downloads_7d momentum fell to 0 per 3d from 7673 (<= 40% of the prior window)
- Accelerating npm_downloads_7d gained 121736 in the last 3d vs 7673 in the 3d before (ratio 15.9x, threshold 1.5x)
- Cooling npm_downloads_7d momentum fell to -131139 per 3d from 84734 (<= 40% of the prior window)
🔎 Known / Unknown
12 of 23 fields unknown
| Machine-callable | — | unknown | — |
| Open source | 1 | inferred | 1 quote(s) |
| Self-hostable | 1 | inferred | 1 quote(s) |
| Bring your own key | — | unknown | — |
| Autonomy level | 2 | inferred | 1 quote(s) |
| Pricing model | — | unknown | — |
| Integrations | vscode | inferred | 1 quote(s) |
“Unknown” means we have not verified it — it is not a “no”. Hard filters never treat unknown as false.
🤖 Agent teardown · skill
Job to be doneTo get instant AI-generated code recommendations and completions while writing code.
AutonomyL2 · tool-calling(evidence: "tool calls, approvals, and follow-up questions run through t…")
Who it is forDevelopers
Prerequisitesopen source · self-hostable
Integrationsvscode
Why it matters
It embeds large models deeply into a mainstream development environment, serving as a prime example of AI agents enabling practical coding workflows.
Evidence quotesverbatim, from the product’s own materials
“public GitHub repository (README fetched)”— structural
“Install the official DeepSeek Harness CLI: npm install -g @deepseek-ai/dsh”— readme
“tool calls, approvals, and follow-up questions run through the official DSH runtime.”— readme
“A VSCode extension that integrates DeepSeek's models”— description
“Extend DSH from the sidebar. Discover and manage Tools, Skills, MCP integrations, Memory, and Agent Hooks loaded by DSH.”— readme
“dsh-vscode gives DSH a Claude Code/Codex-style right sidebar”— readme
Signal source: Show HN
Visit official site →
📛 Official badgefor your site / README
Building DeepSeek Harness in VSCode? Pick a style above — the embed code updates live. Deep color control via URL params: bg= / fg= / accent= (hex). It links back to this page.
On mobile tap Share for WeChat / RED (Xiaohongshu) / X; on desktop use Copy text and paste into the app.
FAQ
What is DeepSeek Harness in VSCode?
DeepSeek Harness in VSCode brings DeepSeek's models directly into the developer's editor to provide real-time code suggestions and completions.
What does DeepSeek Harness in VSCode do?
To get instant AI-generated code recommendations and completions while writing code.
Why does DeepSeek Harness in VSCode matter?
It embeds large models deeply into a mainstream development environment, serving as a prime example of AI agents enabling practical coding workflows.
Is DeepSeek Harness in VSCode open source?
Yes — DeepSeek Harness in VSCode is open source.
Can I self-host DeepSeek Harness in VSCode?
Yes — DeepSeek Harness in VSCode can be self-hosted.
What does DeepSeek Harness in VSCode integrate with?
vscode
How popular is DeepSeek Harness in VSCode?
As tracked by KanonAgent: 2 upvotes (first indexed 2026-08-21).
What are the best DeepSeek Harness in VSCode alternatives?
Similar AI agents tracked by KanonAgent: Stealth Venture, ECC, orca, Agent-Reach, taste-skill, archify.