50fa79407d
Some checks are pending
CI — CoM Config Validation / Validate JSON Configs (push) Waiting to run
CI — CoM Config Validation / Validate YAML Configs (push) Waiting to run
CI — CoM Config Validation / Lint Shell Scripts (push) Waiting to run
CI — CoM Config Validation / Secret Detection (push) Waiting to run
CI — CoM Config Validation / Lint Markdown (push) Waiting to run
CI — CoM Config Validation / Validate CODEOWNERS (push) Waiting to run
Public, sanitized mirror of an AI orchestration command center: agents, skills, MCP servers, slash-command workflows. All infrastructure identifiers, hostnames, mesh IPs/subnets, repo paths, maintainer identity, and hardware fleet specifics scrubbed to <placeholders>; session debug logs and host-specific memory removed. No live credentials. Verified clean by automated leak sweep. See SANITIZATION.md. churchofmalware.org . authorized research only
50 lines
1.7 KiB
Markdown
50 lines
1.7 KiB
Markdown
---
|
|
name: go
|
|
description: Powerhouse task launcher. Loads full context (memory, agents, constitution, TODOs), decomposes task into waves, maps to specialist agents, presents execution plan with approval gate.
|
|
argument-hint: [task description]
|
|
allowed-tools: Bash, Read, Grep, Glob, Agent, WebFetch, WebSearch
|
|
---
|
|
|
|
# /go — CoM Powerhouse Task Launch
|
|
|
|
Before starting any non-trivial task, this command forces a comprehensive context load and decomposition.
|
|
|
|
## Step 1: Context Load
|
|
|
|
Read these files to establish full situational awareness:
|
|
|
|
1. `~/.claude/a2a/agent-cards.json` — Available agents and capabilities
|
|
2. `~/.claude/a2a/constitution.md` — Governance rules and alignment axis (skim Article IV)
|
|
3. Memory files — Current session state, sprint, hardware profile
|
|
|
|
Report a one-line status for each:
|
|
- Enterprise: [X agents across Y pods]
|
|
- Alignment: [Busytown/Stable/Strained/Rapture assessment]
|
|
- Hardware: [oracle or <node>, max concurrent]
|
|
|
|
## Step 2: Task Decomposition
|
|
|
|
Using the task from $ARGUMENTS, invoke Sequential Thinking:
|
|
|
|
1. **Goal:** What is the objective? State it in one sentence.
|
|
2. **Sub-tasks:** Break into discrete, implementable units.
|
|
3. **Dependencies:** Which sub-tasks depend on others?
|
|
4. **Agent mapping:** For each sub-task, identify the best-matching agent from the registry.
|
|
5. **Risk assessment:** What could go wrong? What's the rollback plan?
|
|
|
|
## Step 3: Execution Plan
|
|
|
|
Present the plan with waves, agent assignments, expected outputs, risks, and hardware constraints.
|
|
|
|
## Step 4: Approval Gate
|
|
|
|
Present the plan to the user and ask: **"Launch? [y/N]"**
|
|
|
|
On approval, begin executing Wave 1. On rejection, ask what to modify.
|
|
|
|
## USAGE
|
|
|
|
`/go <task description>`
|
|
|
|
If no arguments provided, ask the user what they want to accomplish.
|