#113113th of 149 in Skills, plugins and rules
observal
Self-hosted registry and telemetry for internal coding agent skills, MCP servers and hooks
- Stars
- 4.3k
- License
- Apache-2.0
- Last commit
- Oct 2026
- Last release
- Oct 2026
- Language
- Python
Missed the latest GitHub check; these facts may be older.
Overview
Observal is a server plus CLI that stores an organization's coding agent resources (agents, skills, MCP servers, hooks, prompts, sandboxes) in one reviewed registry. It generates per-harness config for Claude Code, Cursor, Kiro, Codex, Copilot, OpenCode and others, and captures session traces through installed telemetry hooks. It also publishes approved resources over ARD and lets agents delegate tasks over A2A.
Who it is for: Platform teams managing shared coding agent resources across developers
Strengths
- One install command renders an agent into each supported harness's config format
- Admin review, version diffs and forking of approved releases keep registry changes auditable
- Session traces and usage data feed reports on which agents and tools get used
- Implements ARD v0.91 discovery and A2A v1.0 delegation with team-private visibility rules
Weaknesses
- Needs a multi-service stack (API, web UI, Postgres, ClickHouse) on Docker Engine 24+ with Compose v2
- Telemetry hooks are installed into each developer's harness via the CLI, adding a per-machine setup step
- Value depends on an organization already having many internal agent resources to catalog
- RAM and hardware requirements are not stated in the README
What it needs
- Docker + Compose
- Needs Docker Engine 24+ with Compose v2, PostgreSQL, ClickHouse
- Models: Claude Code, Cursor, Kiro, Codex, GitHub Copilot
Also in Skills, plugins and rules
See all 149| Rank | Project | Score |
|---|---|---|
| 1 | cavemanCuts token use in coding agents with terse replies and compressed inputs | 95 out of 100 |
| 2 | skillsAgent skills for planning, TDD, debugging and code review | 94 out of 100 |
| 3 | eccAgents, skills, hooks and rules bundle for Claude Code, Codex and other harnesses | 94 out of 100 |
| #4 | impeccableDesign skill with 24 commands and a detector for AI-generated frontends | 94 out of 100 |
| #5 | ponytailSkill prompt that makes coding agents write less code and add tests | 92 out of 100 |