metaswarmdsifry/metaswarm
Splits a software development task into multiple specialized AI roles and drives them through a fixed process,…
No ratings yet
Rate this skill
Stars420
Forks54
Watchers420
Open Issues22
InstallOne-line install
UnconfirmedWhether this repository follows the Agent Skills spec (that is, whether it contains a SKILL.md) has not been verified yet. Until that is settled, clone it and use what you need:
git clone https://github.com/dsifry/metaswarm.git
OverviewOverview
Key Features
Splits a software development task into multiple specialized AI roles and drives them through a fixed process, so an agent team divides work, reviews, and delivers like a real engineering team. Key capabilities include: 1. 18 specialized agent personas: ships 18 role personas such as Researcher, Architect, Coder, and Security Audit, switching responsibilities by task phase so one agent does not both design and implement; 2. A structured 9-phase workflow: breaks development into 9 phases (research, design, implementation, review, etc.) and enforces order, preventing agents from skipping steps or missing key gates; 3. Parallel Design Review Gate: launches multi-role design reviews in parallel before coding and blocks later phases if they fail, surfacing architecture issues early; 4. Recursive orchestration: lets a lead agent delegate sub-tasks to lower-level agents and collect results, supporting large task decomposition; 5. Git-native task tracking: writes task state directly into the Git repository so progress and code changes share one source of truth; 6. Knowledge base and quality rubrics: accumulates project context and scores output with rubrics, reducing repeated research and subjective acceptance; 7. External AI tool delegation: hands specific sub-tasks to external AI tools and retrieves their results; 8. Visual review and PR lifecycle automation: visually checks UI output and automates PR creation, review, and merge flow.
Use Cases
medium-to-large development tasks split across multiple AI agents, pre-coding architecture and security review, long-running project task status tracking, PR submission and review automation, visual acceptance of UI changes.
Target Roles
AI / Algorithm Engineer, Full-stack / Indie Developer, DevOps / Operations Engineer, Test / QA Engineer.
RatingUser ratings
Nobody has rated this skill yet. Yours would be the first on the site — your vote becomes the initial average.
RepositoryRepository
| Skill ID | 1971 |
|---|---|
| Data source | GitHub |
| Repository | dsifry/metaswarm |
| Category | AI & Agent Development / Multi-Agent Collaboration & Workflow Orchestration |
| Primary language | Shell |
| License | MIT License |
| Default branch | main |
| GitHub repository ID | 1147763519 |
| Node ID | R_kgDORGl7Pw |
| Rank | #1479 (position within the source ranking) |
| Site-wide popularity rank | #1670 (computed live from Stars, across all 5833 entries) |
GrowthStar / Fork growth
One snapshot per day, tracking how Stars and Forks changed
- Now420
- Previous point · 09-27 421
- vs previous point -1
- 2 data points 14 days
- Now54
- Previous point · 09-27 54
- vs previous point ±0
- 2 data points 14 days
TimelineTimeline
Repository created2026-02-02 07:13:329 months ago
Data last updated2026-10-26 12:14:5723 days ago
Last commit2026-06-19 22:25:295 months ago
CommitLatest commit
Version [v0.12.0] committed by dsifry. Update Category: Feature Addition & Enhancement. Commit details: docs: remove personas deck from this repo. This update introduces new capabilities and module extensions.