No red flags found in any of the 11 categories — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →
by Birfy · Agent Tool · ★ 236
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
🔒 Is agentdescent safe to install? View the security audit →
AgentDescent Gradient descent — but the parameters are agents. A parallel, asynchronous framework for self-evolving agents (skills, prompts, harnesses) where diffs are the gradients and the aggregator is the optimizer. N workers propose edits to a shared artifact in parallel; a barrier-free aggregator merges them into one version-controlled library. Serial self-improvement is bounded at one accepted change per iteration, and merging concurrent edits is the attempt to lift that bound. The place the analogy has to break is the whole design: gradients add, diffs do not. So aggregation is not averaging but conflict resolution, fusion, statistical acceptance and a transactional commit. What it measures Every row names the setting that produced it, and [the full results page](ht
| Stars | 236 |
| Forks | 19 |
| Language | Python |
| Category | Agent Tool |
| License | MIT |
| Quality Score | 66.6449813694934/100 |
| Open Issues | 8 |
| Last Updated | 2026-09-17 |
| Created | 2026-07-24 |
| Platforms | python |
| Est. Tokens | ~17k |
These tools work well together with agentdescent for enhanced workflows:
Looking for a agentdescent alternative? If you're comparing agentdescent with other agent tool tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
Open survey and evidence map for AI agent evolution, self-evolving agents, memory, skills, harnesses, benchmar
Open-source autoresearch powered by autonomous coding agents. Run Claude Code, OpenCode, and Codex with gradin
Model-agnostic plug-n-play LangChain/LangGraph agents powered entirely by MCP tools over HTTP/SSE.
Explore the must-have external ready-made toolkits to integrate with your AI agents built with Python and Type
Debug your AI agents
Easily create LLM tools and agents using plain Bash/JavaScript/Python functions.
Explore other popular agent tool tools:
agentdescent is 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.. It is categorized as a Agent Tool with 236 GitHub stars.
agentdescent is primarily written in Python. It covers topics such as agent-framework, agents, ai-agents.
You can find installation instructions and usage details in the agentdescent GitHub repository at github.com/Birfy/agentdescent. The project has 236 stars and 19 forks, indicating an active community.
agentdescent is released under the MIT license, making it free to use and modify according to the license terms.
The top alternatives to agentdescent on Agent Skills Hub include awesome-agent-evolution, CORAL, DeepMCPAgent. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: