Skip to content
Agentmart
Submit a tool

agent-registry

Dynamic agent registry via js-doc-store-server.

Skill by MauricioPerera

IndexedMIT
Write a review

Chalk Labs

Need help deploying agent-registry?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/mauricioperera-agent-registry

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Dynamic agent registry via js-doc-store-server. Store, version, and discover subagent definitions in a database instead of static markdown files.

Use cases: documents, backend, agent-building

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/agent-registry

What's inside

Skill name
agent-registry
Description
Dynamic agent registry via js-doc-store-server. Store, version, and discover subagent definitions in a database instead of static markdown files.
License
MIT
Instructions
143 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Agent Registry — Dynamic Agent Discovery

This skill extends the subagent pattern with a **hybrid discovery model**: agents live both on the filesystem (`~/.pi/agent/agents/*.md`) and in a `js-doc-store-server` table. When a subagent task runs, pi discovers agents from **both sources** and merges them.

## Problem with Static Filesystem Agents

| Issue | Consequence |
|-------|-------------|
| Agents scattered across machines | No portability between workspaces |
| No versioning | Agent definitions evolve but files don't track it |
| No dynamic updates | Must restart pi to see new agents |
| No sharing | Can't reuse agent definitions across projects |

## The Registry Pattern

```
Filesystem (~/.pi/agent/agents/)
├── scout.md
├── planner.md
├── reviewer.md
└── worker.md

js-doc-store-server (localhost:3000)
└── table: agents
    ├── scout     → content: full markdown
    ├── planner   → content: full markdown
    ├── reviewer  → content: full markdown
    ├── worker    → content: full markdown
    └── custom    → user-defined agents
```

## Schema

```json
{
  "tableName": "agents",
  "columns": [
    { "name": "name",        "type": "text", "required": true },
    { "name": "description", "type": "text" },
    { "name": "version",     "type": "text" },
    { "name": "tags",        "type": "text" },
    { "name": "model",       "type": "text" },
    { "name": "tools",       "type": "text" },
    { "name": "content",     "type": "text", "required": true },
    { "name": "source",      "type": "text" }
  ]
}
```

## Agent Markdown Format

The `content` field stores the **complete markdown** including frontmatter:

```markdown
---
name: scout
description: Fast codebase reconnaissance
tools: read, grep, find, ls, bash
model: qwen3.6:latest
---

You are a scout. Quickly investigate a codebase and return structured findings...
```

## How Discovery Works

When the `subagent` tool executes:

1. **Filesystem scan** — loads `~/.pi/agent/agents/*.md` (user-level)
2. **Project scan** — loads `.pi/agents/*.md` (project-level, if scope allows)
3. **Server query** — `POST /admin/query { tableName: "agents", filter: {} }`
4. **Merge** — server agents overlay filesystem agents by name (server wins)

Server agents have `source: "registry"` and `filePath: "registry://{name}"`.

## Workflow: Register a New Agent

```typescript
// 1. Discover by topic (optional)
const results = await arch_query({
  tableName: "agents",
  filter: { tags: { $regex: "review" } }
});

// 2. Register new agent from conversation
await arch_insert({
  tableName: "agents",
  data: {
    name: "security-auditor",
    description: "Audits code for security vulnerabilities",
    version: "1.0.0",
    tags: "security,audit,subagent",
    model: "qwen3.6:latest",
    tools: "read, grep, find, ls, bash",
    content: `---\nname: security-auditor\n...\n---\n\nYou are a security auditor...`,
    source: "registry"
  }
});

// 3. Verify
const all = await arch_query({ tableName: "agents", filter: {} });
```

## Tool Combinations for Agent Management

| Task | Tool Sequence |
|------|---------------|
| **Discover by topic** | `arch_query` → filter `tags: { $regex: "topic" }` |
| **Load full agent** | `arch_query` → filter `name: "exact-name"` → read `.content` |
| **Register new agent** | `arch_insert` into `agents` table |
| **Update agent version** | `arch_update` with `$set: { version, content }` |
| **List all agents** | `arch_query` with empty filter |
| **Delete obsolete agent** | `arch_remove` with filter `{ name: "agent-name" }` |

## Server-Side Fallback

If the server is unreachable or the token is expired, discovery falls back silently to filesystem-only agents. No error is thrown — the subagent continues working with whatever agents are available.

## Benefits

From MauricioPerera/pi-extension-data-architect, MIT.

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
May 12, 2026 (5 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from MauricioPerera

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get agent-registry running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying agent-registry?

CallGet help
Cite this page

agent-registry skill. Agent Mart. https://agentmart.dev/skills/mauricioperera-agent-registry (last updated 2026-10-24)