Tencent/LoopForge
Self-hosted workflows for coding-agent delivery. Orchestrate CodeBuddy, Codex and Claude Code with scripts and human approval gates. Includes Local workflows for coding assistants.
Read original source
Source record
Recorded metrics
daily_stars · 33 stars today
2026-11-15 04:11:08.796Z
GitHub daily Trending window; attention only
Project analysis
Problem & scenario
LoopForge addresses inconsistent delivery by coding agents through repeatable workflows with review and approval. The supplied evidence names LoopForge; its relationship to WeWorkBuddy is unverified. Example scenario: a developer asks an agent to fix a bug, then manually checks whether planning, tests and review happened. With LoopForge, the requirement follows an SOP containing agent roles, scripts and a human gate, producing artifacts and a delivery history. This makes required stages explicit and gives reviewers a place to return work upstream. The trade-off is workflow configuration and human review effort; the evidence does not establish reduced delivery time.
Features & advantages
LoopForge adds orchestration around named coding-agent CLIs, including Codex and Claude Code: separate roles, Docker workspaces, script steps, approval gates and resumable history. Its connection to WeWorkBuddy is unverified. On a bug-fix task, the documented distinction is scope: a CLI performs agent work, while LoopForge coordinates implementation, review, deterministic checks and sign-off across stages. This can suit teams that want a reusable delivery process while retaining executor choice. The cost is operating another platform and configuring roles and credentials. The sources provide no independent feature comparison with standalone Codex or Claude Code, so they do not establish exclusive capabilities or a measured advantage.
Technical approach
The README documents an SOP execution path, but supplies no commit-pinned code trace; the project is LoopForge, with no verified link to WeWorkBuddy. Triggering the demo SOP runs a shell script that writes a report, then pauses at a human gate. Approval resumes execution to the final artifact. More generally, nodes declare dependencies and artifacts, and the engine starts a step after its dependencies complete. Default Compose deployment creates MySQL and platform services; SOP and conversation containers use the host Docker socket. This enables sandbox creation but grants control of the host daemon, making the documented setup suitable for trusted teams rather than untrusted tenants. Function-level dispatch and persistence behavior remain unverified.
Business model & buyers
The evidence supports a self-hosted software offering, with likely demand from teams seeking consistent agent delivery and human sign-off. This inference concerns LoopForge; its relationship to WeWorkBuddy is unverified. The README states MIT licensing and requires model credentials for agent work; its script-and-approval demo needs no API key. It supplies no official paid plan, subscription price or payment terms. A plausible purchase trigger for associated services would be the burden of deploying, configuring and maintaining shared workflows, but no such commercial offer is documented. Buyers must account for infrastructure, model usage and operating effort; the sources do not quantify those costs or identify paying customers.
Growth & distribution
LoopForge has a documented first-value path: run a script-and-approval demo without model credentials, then configure credentials and repositories for real delivery. Its relationship to WeWorkBuddy is unverified. S1 records GitHub Trending as a discovery source, and the README offers a product tour and installation instructions; no observation date or acquisition volume is supplied. The demo lets users experience an artifact and approval gate before configuring an agent. A plausible adoption path is demo → repository task → reusable team SOPs. Docker installation, credentials and trusted-team access boundaries can slow activation. Repeated SOP use could support retention, but neither the listing nor repository badges establish usage growth or retention.