You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
<em>Plugins distributed as standalone repos under <ahref="https://github.com/agent-sh">agent-sh</a> org — agentsys is the marketplace & installer</em>
23
+
<em>Plugins distributed as standalone repos under <ahref="https://github.com/agent-sh">agent-sh</a> org - agentsys is the marketplace & installer</em>
24
24
</p>
25
25
26
26
<palign="center">
@@ -38,14 +38,14 @@
38
38
39
39
---
40
40
41
-
AI models can write code. That's not the hard part anymore. The hard part is everything around it — task selection, branch management, code review, artifact cleanup, CI, PR comments, deployment. **AgentSys is the runtime that orchestrates agents to handle all of it**— structured pipelines, gated phases, specialized agents, and persistent state that survives session boundaries.
41
+
AI models can write code. That's not the hard part anymore. The hard part is everything around it - task selection, branch management, code review, artifact cleanup, CI, PR comments, deployment. **AgentSys is the runtime that orchestrates agents to handle all of it**- structured pipelines, gated phases, specialized agents, and persistent state that survives session boundaries.
42
42
43
43
---
44
44
> Building custom skills, agents, hooks, or MCP tools? [agnix](https://github.com/agent-sh/agnix) is the CLI + LSP linter that catches config errors before they fail silently - real-time IDE validation, auto suggestions, auto-fix, and 399 rules for Claude Code, Codex, OpenCode, Cursor, Kiro, Copilot, Gemini CLI, Cline, Windsurf, Roo Code, Amp, and more.
45
45
46
46
## What This Is
47
47
48
-
An agent orchestration system — 19 plugins, 49 agents (39 file-based + 10 role-based specialists in audit-project), and 40 skills that compose into structured pipelines for software development. Each plugin lives in its own standalone repo under the [agent-sh](https://github.com/agent-sh) org. agentsys is the marketplace and installer that ties them together.
48
+
An agent orchestration system - 19 plugins, 49 agents (39 file-based + 10 role-based specialists in audit-project), and 40 skills that compose into structured pipelines for software development. Each plugin lives in its own standalone repo under the [agent-sh](https://github.com/agent-sh) org. agentsys is the marketplace and installer that ties them together.
49
49
50
50
Each agent has a single responsibility, a specific model assignment, and defined inputs/outputs. Pipelines enforce phase gates so agents can't skip steps. State persists across sessions so work survives interruptions.
51
51
@@ -57,8 +57,8 @@ The system runs on Claude Code, OpenCode, Codex CLI, Cursor, and Kiro. Install v
57
57
58
58
**Code does code work. AI does AI work.**
59
59
60
-
-**Detection**: regex, AST analysis, static analysis—fast, deterministic, no tokens wasted
61
-
-**Judgment**: LLM calls for synthesis, planning, review—where reasoning matters
-**Judgment**: LLM calls for synthesis, planning, review - where reasoning matters
62
62
-**Result**: 77% fewer tokens for [/drift-detect](#drift-detect) vs multi-agent approaches, certainty-graded findings throughout
63
63
64
64
**Certainty levels exist because not all findings are equal:**
@@ -442,7 +442,7 @@ If something can't be fixed, the workflow replies explaining why and resolves th
442
442
443
443
### /deslop
444
444
445
-
**Purpose:** Finds AI slop—debug statements, placeholder text, verbose comments, TODOs—and removes it.
445
+
**Purpose:** Finds AI slop - debug statements, placeholder text, verbose comments, TODOs - and removes it.
446
446
447
447
**How detection works:**
448
448
@@ -679,7 +679,7 @@ Backed by [agent-analyzer](https://github.com/agent-sh/agent-analyzer) Rust bina
679
679
680
680
### /sync-docs
681
681
682
-
**Purpose:** Sync documentation with actual code changes—find outdated refs, update CHANGELOG, flag stale examples.
682
+
**Purpose:** Sync documentation with actual code changes - find outdated refs, update CHANGELOG, flag stale examples.
683
683
684
684
**The problem it solves:**
685
685
@@ -1093,7 +1093,7 @@ Same principle as good code: single responsibility. The exploration-agent explor
1093
1093
1094
1094
**2. Pipeline with gates, not a monolith**
1095
1095
1096
-
Same principle as DevOps. Each step must pass before the next begins. Can't push before review. Can't merge before CI passes. Hooks enforce this—agents literally cannot skip phases.
1096
+
Same principle as DevOps. Each step must pass before the next begins. Can't push before review. Can't merge before CI passes. Hooks enforce this - agents literally cannot skip phases.
<title>AgentSys - Agent Runtime & Orchestration System</title>
7
-
<metaname="description" content="A modular runtime and orchestration system for AI agents. 19 plugins, 49 agents, 40 skills — structured pipelines for Claude Code, OpenCode, Codex CLI, Cursor, and Kiro.">
7
+
<metaname="description" content="A modular runtime and orchestration system for AI agents. 19 plugins, 49 agents, 40 skills - structured pipelines for Claude Code, OpenCode, Codex CLI, Cursor, and Kiro.">
Copy file name to clipboardExpand all lines: site/ux-spec.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -650,7 +650,7 @@ This disables:
650
650
### Head Content
651
651
```html
652
652
<title>AgentSys - Agent Runtime & Orchestration System</title>
653
-
<metaname="description"content="A modular runtime and orchestration system for AI agents. 19 plugins, 49 agents, 40 skills — structured pipelines for Claude Code, OpenCode, Codex CLI, Cursor, and Kiro.">
653
+
<metaname="description"content="A modular runtime and orchestration system for AI agents. 19 plugins, 49 agents, 40 skills - structured pipelines for Claude Code, OpenCode, Codex CLI, Cursor, and Kiro.">
0 commit comments