Market intelligence
As of 2026-11-21, AgentSidekick indexes 73,820 active agent products (9,469 with full profiles) DER. MCP servers are 93.6% of them, and 52,150 MCP servers were first published in the twelve complete months before this one SRC; 60.4% of all MCP servers are remote-hosted. All figures below come from one ingest snapshot, so they describe levels and launch dates, not growth in stars or downloads.
Launches per month
Counts by month of first publication. Registry records use the MCP Registry publish date; GitHub-discovered records use the repository creation date, which can predate the product. The current month is partial.
MCP servers per monthSRC
Last 6 months: 2026-06 to 2026-11. Whole catalogue, active records. 2026-11 is partial.
- MCP Registry (publish date)
- GitHub-discovered (repo creation date)
Show the data as a table
| MCP Registry (publish date) (servers) | GitHub-discovered (repo creation date) (servers) | |
|---|---|---|
| 2026-06 | 3,706 | 28 |
| 2026-07 | 4,943 | 20 |
| 2026-08 | 6,393 | 41 |
| 2026-09 | 11,693 | 25 |
| 2026-10 | 16,029 | 42 |
| 2026-11 | 15,644 | 18 |
MCP servers per monthSRC
Last 12 months: 2025-12 to 2026-11. Whole catalogue, active records. 2026-11 is partial.
- MCP Registry (publish date)
- GitHub-discovered (repo creation date)
Show the data as a table
| MCP Registry (publish date) (servers) | GitHub-discovered (repo creation date) (servers) | |
|---|---|---|
| 2025-12 | 268 | 23 |
| 2026-01 | 369 | 38 |
| 2026-02 | 1,005 | 58 |
| 2026-03 | 1,828 | 73 |
| 2026-04 | 2,470 | 45 |
| 2026-05 | 2,823 | 35 |
| 2026-06 | 3,706 | 28 |
| 2026-07 | 4,943 | 20 |
| 2026-08 | 6,393 | 41 |
| 2026-09 | 11,693 | 25 |
| 2026-10 | 16,029 | 42 |
| 2026-11 | 15,644 | 18 |
MCP servers per monthSRC
All: 2009-08 to 2026-11. Whole catalogue, active records. 2026-11 is partial.
- MCP Registry (publish date)
- GitHub-discovered (repo creation date)
Show the data as a table
| MCP Registry (publish date) (servers) | GitHub-discovered (repo creation date) (servers) | |
|---|---|---|
| 2009-08 | — | 1 |
| 2013-09 | — | 1 |
| 2014-05 | — | 2 |
| 2015-01 | — | 1 |
| 2015-02 | — | 1 |
| 2015-07 | — | 1 |
| 2016-01 | — | 2 |
| 2016-03 | — | 1 |
| 2016-09 | — | 1 |
| 2016-10 | — | 1 |
| 2016-11 | — | 2 |
| 2016-12 | — | 2 |
| 2017-01 | — | 2 |
| 2017-02 | — | 1 |
| 2017-04 | — | 1 |
| 2017-06 | — | 1 |
| 2017-07 | — | 1 |
| 2017-09 | — | 1 |
| 2018-07 | — | 1 |
| 2018-08 | — | 2 |
| 2019-06 | — | 1 |
| 2019-09 | — | 1 |
| 2019-10 | — | 1 |
| 2019-12 | — | 1 |
| 2020-01 | — | 1 |
| 2020-02 | — | 1 |
| 2020-03 | — | 1 |
| 2020-04 | — | 1 |
| 2020-05 | — | 4 |
| 2020-06 | — | 1 |
| 2020-07 | — | 2 |
| 2020-10 | — | 2 |
| 2020-12 | — | 1 |
| 2021-01 | — | 2 |
| 2021-02 | — | 1 |
| 2021-03 | — | 1 |
| 2021-04 | — | 2 |
| 2021-05 | — | 1 |
| 2021-08 | — | 1 |
| 2021-09 | — | 1 |
| 2021-10 | — | 2 |
| 2021-11 | — | 1 |
| 2021-12 | — | 1 |
| 2022-02 | — | 1 |
| 2022-03 | — | 1 |
| 2022-04 | — | 2 |
| 2022-05 | — | 1 |
| 2022-07 | — | 1 |
| 2022-08 | — | 1 |
| 2023-04 | — | 1 |
| 2023-05 | — | 3 |
| 2023-06 | — | 4 |
| 2023-07 | — | 2 |
| 2023-08 | — | 4 |
| 2023-09 | — | 1 |
| 2023-11 | — | 2 |
| 2023-12 | — | 2 |
| 2024-01 | — | 5 |
| 2024-02 | — | 2 |
| 2024-03 | — | 1 |
| 2024-04 | — | 2 |
| 2024-05 | — | 5 |
| 2024-06 | — | 5 |
| 2024-08 | — | 5 |
| 2024-09 | — | 1 |
| 2024-10 | — | 1 |
| 2024-11 | — | 10 |
| 2024-12 | — | 28 |
| 2025-01 | — | 17 |
| 2025-02 | — | 33 |
| 2025-03 | — | 88 |
| 2025-04 | — | 83 |
| 2025-05 | — | 41 |
| 2025-06 | — | 37 |
| 2025-07 | — | 34 |
| 2025-08 | — | 22 |
| 2025-09 | 436 | 21 |
| 2025-10 | 286 | 20 |
| 2025-11 | 166 | 29 |
| 2025-12 | 268 | 23 |
| 2026-01 | 369 | 38 |
| 2026-02 | 1,005 | 58 |
| 2026-03 | 1,828 | 73 |
| 2026-04 | 2,470 | 45 |
| 2026-05 | 2,823 | 35 |
| 2026-06 | 3,706 | 28 |
| 2026-07 | 4,943 | 20 |
| 2026-08 | 6,393 | 41 |
| 2026-09 | 11,693 | 25 |
| 2026-10 | 16,029 | 42 |
| 2026-11 | 15,644 | 18 |
The earliest MCP Registry publish date on record is in 2025-09; before the registry existed, MCP servers appear only through GitHub repository creation dates, so the step-up is partly a change in what is recorded, not only in what launched.
Other product types per monthSRC
Last 6 months: 2026-06 to 2026-11. The four largest dated types other than MCP servers. 2026-11 is partial.
- Agent skills
- Infrastructure
- Plugins
- Rules packs
Show the data as a table
| Agent skills (products) | Infrastructure (products) | Plugins (products) | Rules packs (products) | |
|---|---|---|---|---|
| 2026-06 | 49 | 18 | 19 | 21 |
| 2026-07 | 39 | 17 | 17 | 10 |
| 2026-08 | 41 | 22 | 11 | 11 |
| 2026-09 | 40 | 12 | 24 | 15 |
| 2026-10 | 58 | 10 | 61 | 15 |
| 2026-11 | 13 | 4 | 9 | 8 |
Other product types per monthSRC
Last 12 months: 2025-12 to 2026-11. The four largest dated types other than MCP servers. 2026-11 is partial.
- Agent skills
- Infrastructure
- Plugins
- Rules packs
Show the data as a table
| Agent skills (products) | Infrastructure (products) | Plugins (products) | Rules packs (products) | |
|---|---|---|---|---|
| 2025-12 | 25 | 3 | 12 | 5 |
| 2026-01 | 72 | 9 | 29 | 7 |
| 2026-02 | 63 | 9 | 27 | 27 |
| 2026-03 | 85 | 21 | 42 | 35 |
| 2026-04 | 95 | 15 | 36 | 26 |
| 2026-05 | 41 | 19 | 17 | 24 |
| 2026-06 | 49 | 18 | 19 | 21 |
| 2026-07 | 39 | 17 | 17 | 10 |
| 2026-08 | 41 | 22 | 11 | 11 |
| 2026-09 | 40 | 12 | 24 | 15 |
| 2026-10 | 58 | 10 | 61 | 15 |
| 2026-11 | 13 | 4 | 9 | 8 |
Other product types per monthSRC
All: 2010-11 to 2026-11. The four largest dated types other than MCP servers. 2026-11 is partial.
- Agent skills
- Infrastructure
- Plugins
- Rules packs
Show the data as a table
| Agent skills (products) | Infrastructure (products) | Plugins (products) | Rules packs (products) | |
|---|---|---|---|---|
| 2010-11 | — | 1 | — | — |
| 2011-04 | 1 | — | — | — |
| 2011-10 | 1 | — | — | — |
| 2013-04 | — | 1 | — | — |
| 2013-06 | — | — | 1 | — |
| 2013-08 | — | 1 | — | — |
| 2014-05 | — | 1 | — | — |
| 2015-09 | — | 1 | — | — |
| 2015-11 | — | — | 1 | — |
| 2016-03 | — | 1 | — | — |
| 2016-05 | — | 1 | — | — |
| 2016-06 | — | 1 | — | — |
| 2016-09 | — | 1 | — | — |
| 2016-10 | 1 | — | — | — |
| 2017-06 | — | 1 | — | — |
| 2017-11 | 1 | — | — | — |
| 2018-04 | 1 | 1 | — | — |
| 2018-11 | — | 1 | — | — |
| 2019-01 | — | — | — | 1 |
| 2019-03 | 1 | — | — | — |
| 2019-04 | — | 1 | — | — |
| 2019-05 | 2 | 1 | — | — |
| 2019-06 | — | 2 | — | — |
| 2019-07 | 1 | — | — | — |
| 2019-08 | 1 | 2 | — | — |
| 2019-09 | — | 3 | — | — |
| 2020-01 | 1 | — | — | — |
| 2020-02 | — | 1 | — | — |
| 2020-05 | — | 1 | — | — |
| 2020-06 | — | 1 | — | — |
| 2020-09 | 1 | — | — | — |
| 2020-10 | 1 | 2 | — | — |
| 2020-11 | — | 2 | — | — |
| 2020-12 | 1 | — | — | — |
| 2021-02 | — | 2 | — | — |
| 2021-03 | — | 1 | — | — |
| 2021-05 | — | 1 | — | — |
| 2021-08 | — | 1 | — | — |
| 2021-09 | 1 | — | — | — |
| 2021-10 | 1 | 1 | — | — |
| 2021-12 | 1 | 1 | — | — |
| 2022-01 | 1 | — | — | — |
| 2022-02 | 1 | — | — | — |
| 2022-03 | 1 | 1 | — | — |
| 2022-04 | — | 3 | — | — |
| 2022-05 | 1 | 2 | — | — |
| 2022-06 | 2 | — | — | — |
| 2022-07 | — | 1 | — | — |
| 2022-08 | — | 1 | — | — |
| 2022-09 | — | 1 | — | — |
| 2022-10 | — | 2 | — | — |
| 2022-11 | — | 2 | — | — |
| 2023-01 | 1 | — | — | — |
| 2023-02 | — | 3 | — | — |
| 2023-03 | — | 4 | — | — |
| 2023-04 | 3 | 7 | 1 | — |
| 2023-05 | 1 | 8 | — | — |
| 2023-06 | — | 10 | — | — |
| 2023-07 | — | 5 | — | — |
| 2023-08 | — | 7 | — | — |
| 2023-09 | — | 3 | — | — |
| 2023-10 | — | 3 | — | — |
| 2023-11 | — | 2 | — | — |
| 2023-12 | 1 | 7 | — | — |
| 2024-01 | — | 2 | — | — |
| 2024-02 | 1 | 3 | — | — |
| 2024-03 | — | 7 | — | — |
| 2024-04 | 1 | 5 | — | 1 |
| 2024-05 | 1 | 2 | — | — |
| 2024-06 | — | 9 | — | — |
| 2024-07 | — | 3 | — | — |
| 2024-08 | 2 | 3 | 1 | — |
| 2024-09 | — | 5 | — | 2 |
| 2024-10 | — | 4 | — | — |
| 2024-11 | 1 | 4 | — | — |
| 2024-12 | 2 | 6 | — | 2 |
| 2025-01 | — | 3 | — | 3 |
| 2025-02 | 4 | 3 | — | 6 |
| 2025-03 | 2 | 5 | 3 | 13 |
| 2025-04 | 4 | 2 | — | 7 |
| 2025-05 | 2 | 3 | — | 7 |
| 2025-06 | 4 | 7 | 1 | 6 |
| 2025-07 | 11 | 4 | 1 | 2 |
| 2025-08 | 5 | 1 | 5 | 6 |
| 2025-09 | 2 | 2 | 5 | 2 |
| 2025-10 | 34 | 6 | 13 | 3 |
| 2025-11 | 9 | 2 | 10 | 2 |
| 2025-12 | 25 | 3 | 12 | 5 |
| 2026-01 | 72 | 9 | 29 | 7 |
| 2026-02 | 63 | 9 | 27 | 27 |
| 2026-03 | 85 | 21 | 42 | 35 |
| 2026-04 | 95 | 15 | 36 | 26 |
| 2026-05 | 41 | 19 | 17 | 24 |
| 2026-06 | 49 | 18 | 19 | 21 |
| 2026-07 | 39 | 17 | 17 | 10 |
| 2026-08 | 41 | 22 | 11 | 11 |
| 2026-09 | 40 | 12 | 24 | 15 |
| 2026-10 | 58 | 10 | 61 | 15 |
| 2026-11 | 13 | 4 | 9 | 8 |
2,396 active records have no date at all and are not counted, mostly agent skills (1,321): skill folders and plugin manifests carry no publish date.
Catalogue composition
All active records by product type. A table, because one type is most of the catalogue and bars would show one bar and slivers.
| Type | Indexed | Share | Dated | Profiled |
|---|---|---|---|---|
| MCP servers | 69,078 | 93.6% | 69,078 | 5,541 |
| Agent skills | 2,056 | 2.8% | 735 | 1,898 |
| Plugins | 1,088 | 1.5% | 346 | 723 |
| Infrastructure | 384 | 0.52% | 357 | 364 |
| Frameworks | 273 | 0.37% | 261 | 248 |
| Rules packs | 267 | 0.36% | 267 | 101 |
| Agent projects | 244 | 0.33% | 244 | 195 |
| IDE extensions | 148 | 0.20% | 0 | 128 |
| Libraries | 116 | 0.16% | 99 | 105 |
| Platforms | 47 | 0.06% | 3 | 47 |
| AI APIs | 29 | 0.04% | 2 | 29 |
| SaaS products | 26 | 0.04% | 3 | 26 |
| Developer tools | 23 | 0.03% | 16 | 23 |
| Agent SDKs | 15 | 0.02% | 6 | 15 |
| Protocols & standards | 14 | 0.02% | 3 | 14 |
| Automation platforms | 12 | 0.02% | 4 | 12 |
MCP hosting mix over time
Remote = the publisher hosts an endpoint; local = a package you run. Derived from registry metadata (remotes vs packages).
MCP servers by hosting, per publish monthDER
Last 6 months: 2026-06 to 2026-11. 1,697 of 69,078 MCP servers declare neither and are excluded.
- Remote
- Local
- Remote and local
Show the data as a table
| Remote (servers) | Local (servers) | Remote and local (servers) | |
|---|---|---|---|
| 2026-06 | 1,398 | 1,455 | 816 |
| 2026-07 | 2,524 | 1,844 | 496 |
| 2026-08 | 4,290 | 1,741 | 318 |
| 2026-09 | 7,947 | 2,807 | 884 |
| 2026-10 | 11,569 | 3,470 | 788 |
| 2026-11 | 10,808 | 4,362 | 419 |
MCP servers by hosting, per publish monthDER
Last 12 months: 2025-12 to 2026-11. 1,697 of 69,078 MCP servers declare neither and are excluded.
- Remote
- Local
- Remote and local
Show the data as a table
| Remote (servers) | Local (servers) | Remote and local (servers) | |
|---|---|---|---|
| 2025-12 | 64 | 182 | 13 |
| 2026-01 | 85 | 241 | 27 |
| 2026-02 | 261 | 681 | 45 |
| 2026-03 | 474 | 1,213 | 121 |
| 2026-04 | 916 | 1,153 | 372 |
| 2026-05 | 976 | 1,271 | 491 |
| 2026-06 | 1,398 | 1,455 | 816 |
| 2026-07 | 2,524 | 1,844 | 496 |
| 2026-08 | 4,290 | 1,741 | 318 |
| 2026-09 | 7,947 | 2,807 | 884 |
| 2026-10 | 11,569 | 3,470 | 788 |
| 2026-11 | 10,808 | 4,362 | 419 |
MCP servers by hosting, per publish monthDER
All: 2025-09 to 2026-11. 1,697 of 69,078 MCP servers declare neither and are excluded.
- Remote
- Local
- Remote and local
Show the data as a table
| Remote (servers) | Local (servers) | Remote and local (servers) | |
|---|---|---|---|
| 2025-09 | 220 | 192 | 15 |
| 2025-10 | 137 | 119 | 17 |
| 2025-11 | 44 | 107 | 8 |
| 2025-12 | 64 | 182 | 13 |
| 2026-01 | 85 | 241 | 27 |
| 2026-02 | 261 | 681 | 45 |
| 2026-03 | 474 | 1,213 | 121 |
| 2026-04 | 916 | 1,153 | 372 |
| 2026-05 | 976 | 1,271 | 491 |
| 2026-06 | 1,398 | 1,455 | 816 |
| 2026-07 | 2,524 | 1,844 | 496 |
| 2026-08 | 4,290 | 1,741 | 318 |
| 2026-09 | 7,947 | 2,807 | 884 |
| 2026-10 | 11,569 | 3,470 | 788 |
| 2026-11 | 10,808 | 4,362 | 419 |
All time: 41,713 remote (60.4%), 20,838 local (30.2%), 4,830 both. Remote listings are cheap to publish (one URL), so their share overstates how many distinct services exist.
Category sizes
All categoriesArea = indexed products in each level-2 category (whole catalogue). One product can sit in several categories, so areas overlap in reality and do not sum to the catalogue.
Indexed products by categoryDER
44 level-2 categories, calculated 2026-11-21.
Show the data as a table
| products | Share | |
|---|---|---|
| Agent Commerce | 3,621 | 6.9% |
| AI Coding Agents | 3,366 | 6.4% |
| Communication & Messaging | 2,687 | 5.1% |
| Public & Real-world Data | 2,011 | 3.8% |
| Social & Streaming Platforms | 2,001 | 3.8% |
| Marketing | 1,944 | 3.7% |
| Cloud Providers | 1,777 | 3.4% |
| Retrieval & RAG | 1,756 | 3.4% |
| Crypto & Blockchain | 1,614 | 3.1% |
| Files & Documents | 1,613 | 3.1% |
| Finance & Accounting | 1,600 | 3.1% |
| Agent Security | 1,481 | 2.8% |
| Web Data | 1,470 | 2.8% |
| Agent Runtime & Deployment | 1,461 | 2.8% |
| DevOps & SRE | 1,392 | 2.7% |
| Research & Knowledge | 1,279 | 2.4% |
| E-commerce | 1,242 | 2.4% |
| Generative Media | 1,232 | 2.4% |
| Markets & Trading | 1,209 | 2.3% |
| Agent Frameworks & Orchestration | 1,192 | 2.3% |
| Analytics & Operations | 1,118 | 2.1% |
| Models & Gateways | 1,115 | 2.1% |
| Design & Creative | 1,072 | 2.1% |
| Agent Observability | 951 | 1.8% |
| API & MCP Development | 912 | 1.7% |
| Browser Automation & Computer Use | 884 | 1.7% |
| Sales | 813 | 1.6% |
| Databases | 805 | 1.5% |
| HR & Recruiting | 793 | 1.5% |
| Integration & Automation Platforms | 758 | 1.5% |
| App Development | 705 | 1.3% |
| Legal & Compliance | 679 | 1.3% |
| Enterprise Applications | 672 | 1.3% |
| Content Creation | 656 | 1.3% |
| Data Engineering & Analytics | 606 | 1.2% |
| Calendar & Scheduling | 594 | 1.1% |
| Source Control | 549 | 1.1% |
| Notes & Documents | 493 | 0.9% |
| Security Engineering | 453 | 0.9% |
| Testing & QA | 451 | 0.9% |
| Project Management | 410 | 0.8% |
| Customer Support | 358 | 0.7% |
| Devices & IoT | 274 | 0.5% |
| Personal Assistants | 181 | 0.3% |
Crowded and emerging capabilities
Labels for 305 level-4 capabilities, calculated 2026-11-21. Size is a percentile among capabilities; momentum is relative to the whole catalogue.
Rule: crowded = size in the top 20% of capabilities; thin = bottom 30%. Momentum = 50 + 50·tanh(ln(((n+1)/(p+1)) / base)), where n and p are launches in the last 90 days and the 90 before, and base = the catalogue's own ratio (45,454 vs 14,427 launches, base 3.15). 50 means growing exactly as fast as the catalogue; ≥ 62 is "outpacing", ≤ 38 "slowing/lagging". Fewer than 5 launches in both windows: no momentum label. DER
Crowded, outpacing the catalogue 10Crowded 16Crowded, slowing 37Outpacing the catalogue 29Established 45Lagging the catalogue 75Emerging 14Thin 78Empty 1
Largest crowded capabilities
| Capability | Indexed | 90d | Prev 90d | Momentum | Label |
|---|---|---|---|---|---|
| SEO Research | 1,073 | 735 | 197 | 58 | Crowded |
| PDF Processing | 1,068 | 712 | 200 | 56 | Crowded |
| Invoice Processing | 927 | 722 | 141 | 72 | Crowded, outpacing the catalogue |
| Agent Memory | 804 | 312 | 223 | 16 | Crowded, slowing |
| YouTube Content | 697 | 495 | 122 | 62 | Crowded, outpacing the catalogue |
| Code Quality | 596 | 356 | 122 | 46 | Crowded |
| Domains & DNS | 573 | 395 | 94 | 64 | Crowded, outpacing the catalogue |
| Consumer Messaging | 561 | 355 | 102 | 55 | Crowded |
| Cloudflare Platform | 545 | 281 | 120 | 35 | Crowded, slowing |
| Remote MCP Hosting | 515 | 273 | 151 | 25 | Crowded, slowing |
Fastest relative momentum (emerging or outpacing)
| Capability | Indexed | 90d | Prev 90d | Momentum | Label |
|---|---|---|---|---|---|
| Local Search | 49 | 39 | 3 | 91 | Outpacing the catalogue |
| Structured Scraping | 159 | 139 | 15 | 89 | Outpacing the catalogue |
| Payroll & Benefits | 135 | 113 | 13 | 87 | Outpacing the catalogue |
| Form Filling | 25 | 15 | 1 | 87 | Emerging |
| Dashboard Generation | 11 | 7 | 0 | 87 | Emerging |
| Meeting Transcription | 45 | 36 | 4 | 85 | Outpacing the catalogue |
| Candidate Sourcing | 62 | 54 | 7 | 83 | Outpacing the catalogue |
| Browser Agents | 71 | 34 | 4 | 83 | Outpacing the catalogue |
| iPaaS Connectors | 172 | 120 | 18 | 80 | Outpacing the catalogue |
| Performance Monitoring | 26 | 11 | 1 | 78 | Emerging |
Competitive density
Largest level-3 subcategories with at least 3 starred repositories. Concentration is over GitHub stars of whole-repository products: popularity, not market share or revenue.
| Subcategory | Indexed | With repo | Stars | Median stars | Top share | HHI | Label |
|---|---|---|---|---|---|---|---|
| Agent Payments & Monetization | 3,621 | 1,709 | 167k | 0 | 17% | 0.08 | Crowded |
| 1,718 | 834 | 205k | 0 | 30% | 0.14 | Crowded | |
| Marketing Automation | 1,670 | 971 | 210k | 0 | 29% | 0.12 | Crowded, outpacing the catalogue |
| Finance Automation | 1,600 | 578 | 33k | 0 | 33% | 0.16 | Crowded, outpacing the catalogue |
| Context Engineering | 1,466 | 1,292 | 2.2M | 9 | 9% | 0.03 | Crowded, slowing |
| Social Networks | 1,463 | 746 | 222k | 0 | 15% | 0.06 | Crowded, outpacing the catalogue |
| Document Processing | 1,405 | 815 | 678k | 0 | 29% | 0.12 | Crowded |
| Blockchain Data | 1,352 | 828 | 99k | 0 | 22% | 0.09 | Crowded, slowing |
| Edge & Serverless Platforms | 1,280 | 934 | 541k | 0 | 19% | 0.07 | Crowded |
| Retrieval-Augmented Generation | 1,193 | 1,027 | 3.6M | 10 | 6% | 0.02 | Crowded, slowing |
| Public Datasets | 1,126 | 849 | 14k | 0 | 18% | 0.10 | Crowded, slowing |
| Code Review | 1,094 | 946 | 950k | 2 | 35% | 0.24 | Crowded, slowing |
Calculated 2026-11-21. An HHI above 0.25 is conventionally "highly concentrated"; here it usually means one famous repository dominates the stars, not that it dominates usage.
Licence mix
Products by stated licenceSRC
Top 10 of 51 licences among 31,724 products that state one. 42,096 (57.0%) state none, mostly registry entries that do not carry a licence field.
Show the data as a table
| products | |
|---|---|
| MIT | 25,591 |
| Apache-2.0 | 4,575 |
| AGPL-3.0 | 781 |
| GPL-3.0 | 236 |
| BSD-3-Clause | 76 |
| CC-BY-4.0 | 65 |
| ISC | 63 |
| CC0-1.0 | 56 |
| MIT-0 | 51 |
| MPL-2.0 | 46 |
Language mix
Linked repositories by primary languageSRC
Top 12 of 77 languages across 33,334 linked repositories; 5,414 have no detected language. One repository per row, however many products it holds.
Show the data as a table
| repositories | |
|---|---|
| TypeScript | 9,860 |
| Python | 7,996 |
| JavaScript | 6,382 |
| Rust | 896 |
| Go | 860 |
| HTML | 466 |
| Shell | 374 |
| C# | 270 |
| Java | 117 |
| Dockerfile | 98 |
| Swift | 93 |
| PHP | 84 |
Package registries and downloads
Downloads include CI runs, mirrors and bots. They are not users, and a single snapshot says nothing about adoption trends.
Products by package registrySRC
26,510 products declare at least one package; a product can declare several registries.
Show the data as a table
| products | |
|---|---|
| npm | 18,135 |
| pypi | 5,670 |
| oci | 2,006 |
| mcpb | 1,873 |
| nuget | 188 |
| cargo | 155 |
Monthly downloads per package, distribution SRC
Percentiles across packages with a figure, snapshot 2026-11-21. Not a leaderboard: a percentile tells you what a typical package sees.
| Registry | Packages | p10 | p25 | Median | p75 | p90 | p99 | Zero |
|---|---|---|---|---|---|---|---|---|
| npm, last 30 days | 1,194 | 294 | 538 | 1.3k | 4.2k | 17k | 2.9M | 2 |
| PyPI, last 30 days | 800 | 164 | 308 | 686 | 2.4k | 12k | 321k | 0 |
Only products whose declared package resolved on npm or PyPI are counted; the last package-registry run finished as partial, so coverage is incomplete.
Known advisories by severity
Advisories matched to declared packagesSRC
111 advisories across 48 products, observed 2026-11-21.
Show the data as a table
| advisories | |
|---|---|
| Critical | 16 |
| High | 55 |
| Moderate | 30 |
| Low | 10 |
An advisory existing does not mean the listed version is affected: compare the vulnerable range with the version you install. Products without a declared package cannot be matched, so none found is not evidence of safety. AgentSidekick has not security-reviewed any product.
Publisher concentration
| Publisher | Active records | Share |
|---|---|---|
| mrfentmen | 2,740 | 3.7% |
| sadri-dridi | 2,366 | 3.2% |
| pipeworx-io | 1,914 | 2.6% |
| mcp-dir | 1,120 | 1.5% |
| usefulapi.io | 1,095 | 1.5% |
| CloudCorpRecords | 562 | 0.76% |
| theluckystrike | 424 | 0.57% |
| cotlet.dev | 398 | 0.54% |
| Evozim | 375 | 0.51% |
| smithery.ai | 209 | 0.28% |
Counted by records, not by distinct services: a publisher listing one server per customer or region inflates its count.
Platforms named in publisher descriptions
All platformsVendor claims: the publisher's own text names the platform. Not compatibility, and not tested.
Products whose description names each platformSRC
Top 15 platforms, all active records, as of 2026-11-21. The note on each bar gives how many of those have a full profile.
Show the data as a table
| products | |
|---|---|
| Claude Code | 1,377 |
| Codex | 788 |
| Cursor | 772 |
| ChatGPT | 732 |
| OpenCode | 121 |
| Google Antigravity | 116 |
| OpenClaw | 104 |
| Gemini CLI | 91 |
| Devin Desktop (formerly Windsurf) | 81 |
| Claude Desktop | 67 |
| GitHub Copilot | 47 |
| Cline | 35 |
| Kiro | 23 |
| Claude (claude.ai apps) | 18 |
| Lovable | 15 |
What we can't show yet
- Star and download growth. Growth needs at least two metric snapshots; there is one, taken on SRC. Every star and download figure on this site is a level at that date. Deltas will appear after the next ingest.
- Funding, acquisitions and pricing changes. No source for these is connected, so none are tracked. Nothing here should be read as a funding signal.
- Change history. 31 ingestion runs have completed, all on 2026-11-21; launch dates come from sources, but detected changes only start accumulating now. See the monitor.
- Usage. No install, active-user or revenue figure exists for any product. Stars measure attention; downloads include automation.
Data: /api/dataset/segments.json · Method: methodology · Gaps: opportunity radar
Page data last updated .