ThinkFlowLab / agentdescent
Gradient descent, but the parameters are agents — a parallel, asynchronous framework for self-evolving agents (skills, prompts, harnesses). Diffs are the gradients; the aggregator is the optimizer.
Why this job: topics: agent-framework, llm-agents, ai-agents, agents
What it does
Host / MCP default loop is Path A (doctor → plan → start → apply); advanced MCP tools need agentdescent mcp --advanced — see docs/mcp-path-a.md.
Get started
pip install "agentdescent[mcp]>=0.9.0"
agentdescent demo
# a complete evolution, offline, no API key, ~10 s
agentdescent doctor
# which agent CLIs and provider keys this machine has- Maintenance
- Active
- Last commit today
- Stars
- 203
- +3 this week
- Licence
- MIT
- Allows business use
- Downloads for
- None listed
- No release files
- Latest version
- v0.9.0
- Released 3 days ago
- Built with
- Python
- 100% of the code
Is it looked after?
- Contributors
- 4
- Last 90 days
- Issues
- 94% closed
- 19 still open
- Pull requests
- 584 merged
- 7 waiting
- Releases
- 13
- 13 in the last year
- Contributing guide (opens in new tab)
- No security policy
- No code of conduct
Similar repos
| Repo | Stars | Last commit | Licence | Downloads for |
|---|---|---|---|---|
| i3T4AN / KADATH | 228 | 16 days ago | Apache-2.0 | No downloads |
More: languages, topics and project record
Languages
- Python 100%
- JavaScript <1%
- Shell <1%
- HTML <1%
Topics
- agent-framework
- agents
- ai-agents
- asynchronous
- evolutionary-algorithms
- gradient-descent
- llm
- llm-agents
- optimizer
- parallel
- prompt-optimization
- self-evolving-agents
Project record
- JobAgent frameworks
- StatusListed
- Commits1,387
- Forks27
- Watching9
- Stars in 28 days+31
- CreatedJul 24, 2026
- First seen by usNov 7, 2026
- CheckedNov 9, 2026