overstory-integration

Integrates overstory Claude Code agent swarm with nanobot orchestrator. Manages agent lifecycle, SQLite mail bridge, and git worktree coordination.

Safety Notice

This listing is from the official public ClawHub registry. Review SKILL.md and referenced scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "overstory-integration" with this command: npx skills add austindixson/overstory-integration

overstory Integration

Description

Integrates overstory's Claude Code agent swarm system with the nanobot orchestrator. Provides agent lifecycle management, an inter-agent SQLite mail bridge, and git worktree coordination for parallel agent workstreams.

Architecture

┌──────────────┐       ┌───────────────────┐
│   nanobot     │──────▶│ overstory_wrapper │──▶ overstory CLI (tmux sessions)
│  orchestrator │       └───────────────────┘
│               │       ┌───────────────────┐
│               │──────▶│ agent_lifecycle    │──▶ SQLite (agent_lifecycle.db)
│               │       └───────────────────┘
│               │       ┌───────────────────┐
│               │──────▶│   mail_bridge     │──▶ SQLite (.overstory/mail.db)
└──────────────┘       └───────────────────┘
  • overstory_wrapper.py — Thin wrapper around the overstory CLI. Spawns agents via tmux, manages worktrees, installs hooks.
  • agent_lifecycle.py — Tracks agent state (spawned → running → completed/failed/terminated) in a local SQLite database. Provides async monitoring and cleanup.
  • mail_bridge.py — Thread-safe message bridge using overstory's native SQLite mail format. Supports direct messages, broadcasts, and threaded conversations.

Requirements

  • Python 3.9+
  • overstory CLI installed and on PATH (or set OVERSTORY_BIN)
  • git (for worktree management)
  • tmux (overstory uses tmux for agent sessions)
  • No external Python dependencies (stdlib only)

Environment Variables

VariableDefaultDescription
OVERSTORY_BINoverstoryPath to overstory binary
OVERSTORY_WORKSPACEcurrent directoryWorkspace root for overstory operations

Commands

Initialize workspace

python3 scripts/overstory_wrapper.py init --workspace /path/to/project

Spawn an agent

python3 scripts/overstory_wrapper.py sling \
  --task-id TASK-42 \
  --capability code \
  --name builder-alice \
  --description "Implement auth module" \
  --worktree

Check status

python3 scripts/overstory_wrapper.py status
python3 scripts/overstory_wrapper.py status --agent builder-alice --verbose

Inspect agent transcript

python3 scripts/overstory_wrapper.py inspect --agent builder-alice --lines 100

Kill an agent

python3 scripts/overstory_wrapper.py kill --agent builder-alice

Worktree management

python3 scripts/overstory_wrapper.py list-worktrees
python3 scripts/overstory_wrapper.py cleanup-worktree --agent builder-alice

Agent lifecycle

python3 scripts/agent_lifecycle.py spawn --task "Fix login bug" --capability code --name bug-fixer
python3 scripts/agent_lifecycle.py list-active --json
python3 scripts/agent_lifecycle.py wait --agent bug-fixer --timeout 1800
python3 scripts/agent_lifecycle.py cleanup --max-age 24

Mail bridge

python3 scripts/mail_bridge.py send --from orchestrator --to builder-alice --subject "Priority change" --body "Switch to auth module"
python3 scripts/mail_bridge.py read --agent builder-alice --unread
python3 scripts/mail_bridge.py broadcast --from orchestrator --channel @all --message "Stand down"
python3 scripts/mail_bridge.py create-thread --agents orchestrator,builder-alice --subject "Auth design"
python3 scripts/mail_bridge.py reply --thread THREAD-ID --from builder-alice --body "Done with phase 1"

Usage as Python Module

from overstory_wrapper import OverstoryWrapper
from agent_lifecycle import AgentLifecycle
from mail_bridge import MailBridge

wrapper = OverstoryWrapper("/path/to/workspace")
wrapper.init()

lifecycle = AgentLifecycle()
agent = lifecycle.spawn_agent(task="Build API", capability="code", name="api-builder")

bridge = MailBridge("/path/to/workspace")
bridge.send("orchestrator", "api-builder", "Context", "Here are the specs...")

Database Schemas

agent_lifecycle.db (~/.nanobot/agent_lifecycle.db)

ColumnTypeDescription
idINTEGER PRIMARY KEYAuto-increment
agent_nameTEXT UNIQUEAgent identifier
capabilityTEXTAgent capability (code, research, etc.)
statusTEXTspawned/running/completed/failed/terminated
taskTEXTTask description
resultTEXTFinal output (nullable)
start_timeREALUnix timestamp
end_timeREALUnix timestamp (nullable)
timeoutINTEGERMax runtime in seconds

mail.db (.overstory/mail.db)

ColumnTypeDescription
idINTEGER PRIMARY KEYAuto-increment
thread_idTEXTThread identifier (nullable)
from_agentTEXTSender agent name
to_agentTEXTRecipient agent name
subjectTEXTMessage subject
bodyTEXTMessage body
priorityTEXTnormal/high/urgent
readINTEGER0=unread, 1=read
created_atREALUnix timestamp

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Coding

browser-file-uploa

Automates file uploads on web pages via agent-browser CLI or Python script, supporting flexible file paths and optional file input selectors.

Registry SourceRecently Updated
Coding

Rust Code Review

Reviews Rust code for ownership, borrowing, lifetime, error handling, trait design, unsafe usage, and common mistakes. Use when reviewing .rs files, checking...

Registry SourceRecently Updated
Coding

Crxjs

CRXJS Chrome extension development — true HMR for popup, options, content scripts, side panels, manifest-driven builds, dynamic content script imports (`?scr...

Registry SourceRecently Updated
1140samber
Coding

Atlas Framework

ATLAS Framework - Structured AI-assisted development methodology with GOTCHA 6-layer architecture and 5-step app building workflow. Use when building applica...

Registry SourceRecently Updated