Distilly
#13 in Agenten-FrameworksUnknown · 2× · tolest 21. Sept. 2026
37
Momentum
Distilly (originally "Colleague Skill") is an Open Source project on GitHub (titanwings/distilly) that distills a person's experience, judgment, and communication style into reusable "Agent Skills". From source material such as chat messages, documents, interviews, and public data, it generates "Person Profiles" that can be installed and invoked as Agent Skills in supported AI Agent environments. The project covers three person categories: colleagues/employees, relationships/family, and public figures/celebrities. It is free and available under the MIT License, with no separate commercial product website outside of GitHub.
Momentum-Verloop
24.06.22.09.
Features
| Autonomy Level | Not an autonomous agent itself, but a person-modeling layer: generates source-grounded Person Profiles installed as a Skill inside a host agent |
| Channels | Lark, DingTalk, Slack (for material collection); skill execution across eight agent hosts (Claude Code, Hermes, OpenClaw, Codex, DeepSeek Harness, Pi, Grok Build, OpenCode) |
| License | MIT License © titanwings |
| Modalities | Text/documents, chat logs, interviews, subtitles/transcripts (multimodal source materials for profile creation) |
| Platform | GitHub project; native local Skill discovery for Claude Code, Hermes, OpenClaw, Codex, DeepSeek Harness, Pi, Grok Build, and OpenCode |
| Price | Free, open source (no commercial pricing model found) |
| Release Date | March 30, 2026 (repo created); 2026-08-24 update with host support |
| Tool/MCP Integrations | Supports data collection via Lark, DingTalk, and Slack; no dedicated MCP servers documented |