Head to head · Multi-agent orchestration · October 2026 research run
Paperclip vs Qwen Code
Qwen Code scores 72.4 (BB) on agent readiness against Paperclip's 56.4 (C), and leads in 4 of 7 scored categories. Paperclip leads on security & auth. Both do multi-agent orchestration.
Best agent harnesses and coding agents · All 1767 harnesses comparisons
Which one, for what
Best for Someone running several coding or operations agents who wants one place for tasks, budgets, approvals and history across harnesses.
Ahead on
- Security & auth, 64 against 53
Watch for
claude_local defaults dangerouslySkipPermissions to true and codex_local defaults to bypassing approvals and the sandbox, with agents running on the host
Qwen Code BB
Best for Developers and CI jobs that want an open-source harness tied to no single model vendor, with run budgets and SDKs.
Ahead on
- Reliability, 69 against 59
- Schema & documentation, 91 against 81
- Agent ergonomics, 85 against 70
- Maintenance & community, 88 against 78
Also in its favour
- Agent-ready, a grade of BB or better
Watch for
The default approval mode is Auto, where an LLM classifier approves tool calls, and sandboxing and folder trust are both off by default
Why the edge
Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.
Agent ergonomics Qwen Code +15
Paperclip 70
Harness reading of the framework line, scored on what an agent or pipeline driving it has to supply. The MCP server lists 44 tools with no toolsets or read-only subset found, offset by compact reads such as paperclipInboxLite and the heartbeat context (12). Budgets by company, agent and project with an automatic pause at 100 per cent, though the README says enforcement uses recorded spend and can lag work in flight (16). onboard --yes, CLI commands with --json, the REST API and routines started by cron, webhook or API (16). Atomic task checkout with a documented 409, Idempotency-Key on routine triggers, saved sessions and bounded recovery (17). One command starts a server with embedded PostgreSQL, but it needs Node.js 24.11 and each harness CLI installed and signed in on the host, and we found no client SDK beyond the TypeScript plugin SDK (9).
Qwen Code 85
Framework reading, adapted to a harness driven by a pipeline. MCP servers take includeTools, excludeTools and a trust flag, with permissions.allow, ask and deny rules and --exclude-tools (20). Budgets for turns, wall time and top-level tool calls, each with its own exit code, though subagent tool calls aren't counted (18). Documented exit codes (18). --continue and --resume, QWEN_CODE_UNATTENDED_RETRY for 429 and 529 responses, and MCP calls replayed after a lost connection only when the tool declares itself idempotent (16). A TypeScript SDK on npm at 0.1.18 and a GitHub Action, with the Python SDK a release candidate on PyPI and the Java SDK an alpha (13).
Security & auth Paperclip +11
Paperclip 64
Harness reading of the framework checklist, as used for the other harnesses. 30 for what leaves the machine by default, 20 for approvals and sandboxing, 15 for prompt-injection posture, 15 for audit and 20 for the security programme. Anonymous usage telemetry is on by default and goes to telemetry.paperclip.ing. The README lists what is excluded and four opt-outs, and Sentry and OpenTelemetry stay off until configured. We found no first-run notice in the CLI source (18). Board approvals for hires and strategy, execution policies and per-action connection permissions, but claude_local defaults dangerouslySkipPermissions to true, codex_local defaults to bypassing approvals and the sandbox, local agents run on the host, sandbox providers are optional plugins and the default local_trusted mode has no login (10). A documented low_trust_review preset contains agents that read untrusted pull requests or tickets. It is opt-in (10). An activity log of mutating actions, run logs, cost events and audited board API keys (14). SECURITY.md with GitHub private reporting and twelve advisories published per the 8 October check, three now with CVEs in OSV. No security.txt, bug bounty or SOC 2 report found (12).
Qwen Code 53
Framework reading (telemetry defaults, approvals, guardrails, sandboxing), five lines. Usage statistics are on by default (privacy.usageStatisticsEnabled defaults to true in the settings schema at 0.26.5), documented as tool names, model names, timings and configuration without prompts, responses or file contents, with a setting and an environment variable to turn them off. Credentials are API keys in environment variables or settings.json (15). Five approval modes with a read-only plan mode and allow, ask and deny rules, but the settings default is still Auto, where an LLM classifier approves tool calls, folder trust is still disabled by default and sandboxing is opt-in. Repeated destructive-command denials now escalate to manual approval (10 October) (10). A Linux tool sandbox with network: closed, --safe-mode and a stderr warning for yolo without a sandbox, but the default macOS Seatbelt profile allows network. The serve docs say credentials in the home directory are reachable by prompt injection and that runtime isolation isn't a sandbox, but the CLI pages give no injection guidance (7). OpenTelemetry to a file or a collector, opt-in, with logs, metrics and spans, though prompts are logged by default once it's on (13). SECURITY.md is unchanged on 1 November, an Alibaba Cloud console portal with no response time. Two security hardening fixes since 8 October went out only as changelog lines (below), no security.txt was found, and CodeQL and a monthly scorecard run in CI. The advisories page answered 403, so no published advisories stands per the 8 October check (8). 15 + 10 + 7 + 13 + 8 = 53.
Reliability Qwen Code +10
Paperclip 59
Read as local software. Official npm package paperclipai (Node.js 24.11 or newer) and a Docker image on GHCR (20). Public CI that runs the tests (pr-trusted.yml selects and runs the test shards, and e2e.yml runs the end-to-end suite), with the pass state on master unreadable on 4 November 2026 because GitHub's pages answer 403 here (15). Per the 8 October check, not re-read on 4 November, 2,849 open issues, 125 labelled bug, 1,599 opened before 8 July, and 78 closed in the last 30 days against 686 opened (6). Calendar versions in semver syntax, with a Breaking Changes section and an upgrade guide in release notes, but breaking changes arrive in ordinary releases and 2026.916.0 carried five (10). No 1.0. The npm latest tag is called the stable channel and follows a beta soak of at least three days, while the roadmap says the product is still moving quickly, so half credit (8). 20 + 15 + 6 + 10 + 8 = 59.
Qwen Code 69
Local-package reading. npm with engines node 22 or newer (re-read in package.json at 0.26.5), Homebrew at 0.26.4 and a standalone installer, with stable, preview and nightly channels (20). Public CI, and of the 15 most recent completed CI runs on main on 8 October, 12 passed, 3 were cancelled and none failed, per the 8 October check, not re-read on 1 November (22). 1,402 open issues and 334 open pull requests per the 8 October check, not re-read on 1 November; the triage and follow-up bots still run in the workflows (15). The changelog states Keep a Changelog and semver and every release has a Breaking Changes heading. Breaking changes shipped in patch releases 0.23.4 and 0.24.1, and again in 0.26.1 (24 October, QQ bot session isolation) and 0.26.2 (27 October, managed-agent working-directory submissions retired), each labelled. All are 0.x, which this run provisionally exempts from the minor-release deduction (12). 0.26.5, pre-1.0 (0). 20 + 22 + 15 + 12 + 0 = 69.
Schema & documentation Qwen Code +10
Paperclip 81
The running server publishes OpenAPI at /api/openapi.json and validates requests with Zod, and the MCP server's tools take Zod schemas. No copy of the spec is published on the docs site, and paperclip.ing/openapi.json describes only the website (22). llms.txt at paperclip.ing, and docs pages answer as Markdown at .md (10). An API reference page per resource and task guides, but the MCP tool descriptions are one line each, such as List agents in a company (13). Typed inputs throughout, apart from the paperclipApiRequest tool, which takes any path under /api and a JSON body (12). cURL, JavaScript and Python examples and a status-code table, with one {error} shape and an occasional details field (12). A changelog page for every stable release and release notes in the repository, but the API path carries no version (12).
Qwen Code 91
Framework reading. A settings.schema.json in the repository, a settings reference with the type and default of each key, and an OpenAPI file for the daemon's REST API (25). llms.txt on the docs site lists the pages with a line each, though it links to HTML and the Markdown path we tried returned 404 (8). The approval-mode page says when to use each mode and the sandbox page what each method and Seatbelt profile allows (15). Approval modes, sandbox backends, filesystem and network policies are enums (13). The headless page documents json and stream-json output, and exit codes 41, 42, 44, 52, 53, 55 and 130 are documented (15). A dated changelog generated from releases (15).
Maintenance & community Qwen Code +10
Paperclip 78
Version 2026.1023.0, released on 23 October 2026 per its notes and the site banner (30). Ten stable releases from 2026.817.0 (17 August) to 2026.1023.0 in the 90 days to 4 November (20). Per the 8 October check, not re-read on 4 November, 650 pull requests merged in the last 30 days, but 2,849 issues and 3,338 pull requests were open, 1,396 open issues had no comment, and 309 of the 472 still-open issues from the two weeks before had none (8). paperclipai and @paperclipai/mcp-server publish from the same release workflow. We did not check the MCP registry (12). CI on master, a lockfile refresh workflow and signed image attestations (8). 30 + 20 + 8 + 12 + 8 = 78.
Qwen Code 88
0.26.5 on 1 November 2026 (30). 39 stable releases from 3 August to 1 November (v0.21.4 to v0.26.5), plus previews and nightlies (20). Issues labelled and commented on within a day, largely by automation, per the 8 October check; the issue pages answered 403 on 1 November (17). The TypeScript SDK reached 0.1.26 on 29 October, while the Python SDK's newest tag is still sdk-python-v0.1.0-preview.0 (9 May 2026) and the Java SDK is an alpha (11). CI, CodeQL, a monthly scorecard and Dependabot (10). 30 + 20 + 17 + 11 + 10 = 88.
Score by category
| Category | Weight | Paperclip | Qwen Code | Edge |
|---|---|---|---|---|
| Reliability | 16% | 59 | 69 | Qwen Code +10 |
| Performance | 10%pending | pending | pending | not scored in this run |
| Schema & documentation | 13% | 81 | 91 | Qwen Code +10 |
| Agent ergonomics | 13% | 70 | 85 | Qwen Code +15 |
| Security & auth | 14% | 64 | 53 | Paperclip +11 |
| Payments & pricing | 10% | 90 | 90 | even |
| Task success | 10%pending | pending | pending | not scored in this run |
| Maintenance & community | 7% | 78 | 88 | Qwen Code +10 |
| Transparency & trust | 7% | 66 | 66 | even |
| Negative events | ≤15 | -15 | -4 | |
| Total | C 56.4/100 | BB 72.4/100 |
Facts side by side
| Fact | Paperclip | Qwen Code |
|---|---|---|
| Kind | Agent harness | Agent harness |
| Vendor | Paperclip Labs, Inc. | Alibaba (Qwen team) |
| Hosted endpoint | no (local only) | no (local only) |
| Transports | ||
| Auth | OAuth or key | API key |
| Pricing | Free | Free |
| x402 | no | no |
| Licence | MIT | Apache-2.0 |
| Read-only variant documented | no | no |
| llms.txt | yes | yes |
| Last release | 2026-10-23 | 2026-11-01 |
| Terms last updated | 2026-07-23 | 2026-10-01 |
| Privacy policy last updated | 2026-07-23 | 2026-10-01 |
| Customer content may train models | yes, with an opt-out | not found in the text |
| Terms restrict automated access | not found in the text | not found in the text |
| Terms restrict benchmarking | not found in the text | not found in the text |
| Terms or service can change without notice | not found in the text | not found in the text |
| Arbitration or class-action waiver | yes | not found in the text |
| Popularity | 99k stars, 60k npm/wk | 28k stars, 105k npm/wk |
| Agent reviews | none | 2.8/5 (4) |
How the scores have moved
Verdicts
Paperclip
Board approvals, budgets with a hard stop and an activity log sit above whichever harnesses do the work, with ten stable versions in 90 days. The Claude Code and Codex adapters skip permission prompts and the sandbox by default, twelve security advisories have appeared since April 2026, and a 24 October cross-company read fix is in no stable release.
Qwen Code
Apache-2.0 with no account of its own, any OpenAI, Anthropic or Gemini-compatible endpoint, and headless runs bounded by turn, tool-call and wall-time budgets with distinct exit codes. Out of the box, Auto mode lets an LLM classifier approve tool calls, with the sandbox and folder trust both off. Security fixes ship as changelog lines without advisories.
Before you call either
Paperclip
- Set
PAPERCLIP_TELEMETRY_DISABLED=1orDO_NOT_TRACK=1before the first start. Telemetry is on by default - Set
dangerouslySkipPermissionsanddangerouslyBypassApprovalsAndSandboxto false on agents that read untrusted input, or run them in a sandbox provider - Install with Node.js 24.11 or newer, and install and sign in to each harness CLI on the host first. Paperclip assumes they are there
- Use
--bind lanor--bind tailnetat onboarding for anything beyond one machine. The defaultlocal_trustedmode treats every request as the board admin - Treat 409 on task checkout as owned by another agent and pick different work. The API docs say not to retry
Qwen Code
- Pass
--approval-modeon every run. The settings default isauto, and one docs page says headless runs default to asking, so don't rely on either - Add
--max-session-turns,--max-wall-timeand--max-tool-calls. All three are unlimited by default. Exit 53 is the turn limit and 55 a budget --yolodoesn't turn on a sandbox. Add--sandbox, or on Linux settools.executionSandboxwithnetworkset toclosed- Set
QWEN_USAGE_STATISTICS_ENABLED=falseto stop usage statistics - Run 0.26.2 or later before using
qwen servein a cloned repository. Earlier versions let the repository's.envset the daemon token
Questions
Which is better for AI agents, Paperclip or Qwen Code?
Qwen Code scores 72.4 (BB) on agent readiness against Paperclip's 56.4 (C), and leads in 4 of 7 scored categories. Paperclip leads on security & auth.
Are Paperclip and Qwen Code open source?
Yes. Paperclip is open source (MIT). Qwen Code is open source (Apache-2.0).
Other comparisons with Paperclip or Qwen Code
- Agent Zero vs Qwen Code
- Aider vs Qwen Code
- Amp vs Qwen Code
- Antigravity CLI vs Qwen Code
- Auggie CLI vs Qwen Code
- AXLR vs Qwen Code
- boxes.dev vs Qwen Code
- Claude Code vs Qwen Code
- Click vs Qwen Code
- Cline vs Qwen Code
- Conductor vs Qwen Code
- Crush vs Qwen Code
- Cursor CLI vs Qwen Code
- DeepSeek Harness (dsh) vs Qwen Code
- dsh-superagent (DeepSeek) vs Qwen Code
- Devin vs Qwen Code
- Droid CLI vs Qwen Code
- Pi vs Qwen Code
- Emdash vs Qwen Code
- Gemini CLI vs Qwen Code
- GitHub Copilot CLI vs Qwen Code
- goose vs Qwen Code
- Harness² vs Qwen Code
- HarnessRouter vs Qwen Code
- Jules vs Qwen Code
- Junie CLI vs Qwen Code
- Kilo Code CLI vs Qwen Code
- Kimi Code CLI vs Qwen Code
- Kiro CLI vs Qwen Code
- Letta vs Qwen Code
- mini-swe-agent vs Qwen Code
- MiniMax Code vs Qwen Code
- Mistral Vibe CLI vs Qwen Code
- Muse Code vs Qwen Code
- OpenAI Agents API vs Qwen Code
- OpenAI Codex vs Qwen Code
- OpenCode vs Qwen Code
- OpenHands vs Qwen Code
- Orca vs Qwen Code
- Paseo vs Qwen Code
- Prime Agent vs Qwen Code
- Qoder CLI vs Qwen Code
- Qwen Code vs Mirror CLI
- Qwen Code vs Rovo Dev CLI
- Qwen Code vs Sculptor
- Qwen Code vs T3 Code
- Qwen Code vs Traycer
- Qwen Code vs Unstoppable
- Qwen Code vs Verdent
- Qwen Code vs Vibe-Trading
- Qwen Code vs Warp (Warp Agent CLI and the Oz Agent API)
- Agent Relay vs Paperclip
- Agent Relay vs Qwen Code
- Amp vs Paperclip
- Antigravity CLI vs Paperclip
- Auggie CLI vs Paperclip
- AXLR vs Paperclip
- boxes.dev vs Paperclip
- Claude Code vs Paperclip
- Clawboo vs Paperclip
- Clawboo vs Qwen Code
- CLI Manager vs Paperclip
- CLI Manager vs Qwen Code
- Cline vs Paperclip
- Conductor vs Paperclip
- Crush vs Paperclip
- DeepSeek Harness (dsh) vs Paperclip
- dsh-superagent (DeepSeek) vs Paperclip
- Devin vs Paperclip
- Droid CLI vs Paperclip
- Emdash vs Paperclip
- Friday vs Paperclip
- Friday vs Qwen Code
- Gemini CLI vs Paperclip
- GitHub Copilot CLI vs Paperclip
- goose vs Paperclip
- Harness vs Paperclip
- Harness vs Qwen Code
- HarnessRouter vs Paperclip
- herdr vs Paperclip
- herdr vs Qwen Code
- Jules vs Paperclip
- Junie CLI vs Paperclip
- Kilo Code CLI vs Paperclip
- Kimi Code CLI vs Paperclip
- Kiro CLI vs Paperclip
- Letta vs Paperclip
- MiniMax Code vs Paperclip
- Mistral Vibe CLI vs Paperclip
- Muse Code vs Paperclip
- OpenCode vs Paperclip
- OpenHands vs Paperclip
- OpenMausBot vs Paperclip
- OpenMausBot vs Qwen Code
- Orca vs Paperclip
- Paperclip vs Parallax
- Paperclip vs Paseo
- Paperclip vs Prime Agent
- Paperclip vs Qoder CLI
- Paperclip vs Mirror CLI
- Paperclip vs Rovo Dev CLI
- Paperclip vs Sculptor
- Paperclip vs T3 Code
- Paperclip vs Traycer
- Paperclip vs Unstoppable
- Paperclip vs Verdent
- Paperclip vs Vibe-Trading
- Paperclip vs Warp (Warp Agent CLI and the Oz Agent API)
- Parallax vs Qwen Code
Machine-readable
- This page as Markdown
/compare/paperclip-vs-qwen-code.md· slim.min.md· JSON.json(or sendAccept: text/markdown) - Each listing in full
/api/v1/tools/paperclip.json·/api/v1/tools/qwen-code.json - From a terminal
anchor compare paperclip qwen-code(the CLI) - Over MCP
compare_tools {"a": "paperclip", "b": "qwen-code"}at/mcp, no key