boshu2/agentops▌

Quick Ref: One command, full lifecycle. /discovery → /crank → /validation. Thin wrapper that delegates to phase orchestrators. YOU MUST EXECUTE THIS WORKFLOW. Do not just describe it. THREE-PHASE RULE + FULLY AUTONOMOUS. Read references/autonomous-execution.md — it defines the mandatory 3-phase lifecycle, autonomous execution rules, anti-patterns, and phase completion logging. Unless --interactive is set, RPI runs hands-free. Do NOT stop after Phase 2. Do NOT ask the user anything between phases

YOU MUST EXECUTE THIS WORKFLOW. Do not just describe it.

Maintain GOALS.yaml and GOALS.md fitness specifications. Use ao goals CLI for all operations.

Parallel multi-model consensus review with configurable perspectives, debate rounds, and deep exploration. \n \n Supports three task modes: validate (PASS/WARN/FAIL verdicts), brainstorm (explore alternatives), and research (deep analysis with trade-offs) \n Spawns 2–3 independent judges by default; scales to 6+ agents with --deep , --mixed (Claude + Codex), and --explorers sub-agents, capped at 12 total \n Debate mode ( --debate ) runs two-round adversarial review with verdict exchange and posi

Graph-based issue tracker that survives conversation compaction.

Typically runs automatically via SessionStart hook.

Purpose: Provide composable, repeatable security/internal-testing primitives for authorized binaries and repo-managed prompt surfaces.

PR-specific validation that ensures changes are clean, focused, and ready.

Learn from PR outcomes by analyzing accept/reject patterns.

Provide authoritative, current guidance from OpenAI developer docs using the developers.openai.com MCP server. Always prioritize the developer docs MCP tools over web.run for OpenAI-related questions. Only if the MCP server is installed and returns no meaningful results should you fall back to web search.

Purpose: Run repeatable security checks across code, scripts, hooks, and release gates.

Purpose: Detect and auto-fix common skill hygiene issues across the skills/ directory.

Parse AgentOps skills into a universal SkillBundle format, then convert to target agent platforms.

Reverse-engineer a product into a mechanically verifiable feature inventory + registry + spec set, with optional security-audit artifacts and validation gates.

Purpose: Separate WHAT from HOW. Explore the problem space before committing to a solution.

recover

Purpose: One command to pull the latest skills from the repo and install them globally across all agents.

Typically runs automatically via SessionEnd hook.

Language-specific coding standards loaded on-demand by other skills.

One job: Tell a new user what AgentOps does and what to do first. Fast.

This directory contains shared reference documents used by multiple skills:

Track progress through the RPI workflow with permanent gates.

You have access to workflow skills for structured development.

Purpose: Generate a README that converts skimmers into users and satisfies deep readers — then validate it with a council.

Quick Ref: Trace design decisions through CASS sessions, handoffs, git, and artifacts. Output: .agents/research/YYYY-MM-DD-trace-*.md

Purpose: Single-screen overview of your current state. What am I working on? What happened recently? What should I do next?

YOU MUST EXECUTE THIS WORKFLOW. Do not just describe it.

Quick Ref: Execute single issue end-to-end. Output: code changes + commit + closed issue.

Quick Ref: Create structured handoff for session continuation. Output: .agents/handoff/YYYY-MM-DD-<topic>.md + continuation prompt.

Purpose: Is this code ready to ship?

Purpose: Wrap up completed work — validate it shipped correctly, extract learnings, process the knowledge backlog, activate high-value insights, and retire stale knowledge.

Quick Ref: Decompose goal into trackable issues with waves. Output: .agents/plans/*.md + bd issues.

YOU MUST EXECUTE THIS WORKFLOW. Do not just describe it.

Measure what's wrong. Fix the worst thing. Measure again. Compound.

Quick Ref: Deep codebase exploration with multi-angle analysis. Output: .agents/research/*.md

Purpose: Guide the user through creating a PRODUCT.md that unlocks product-aware reviews in /pre-mortem and /vibe, including the default quick-mode inline paths.

The lead orchestrates, Codex agents execute. Each agent gets one focused task. The team lead prevents file conflicts before spawning — the orchestrator IS the lock manager.

Spawn isolated agents for parallel task execution with runtime-native team selection and fresh context per wave. \n \n Auto-selects spawn backend based on runtime: Claude Native Teams in Claude sessions, Codex sub-agents in Codex sessions, or inline fallback if multi-agent unavailable \n Executes tasks in dependency-ordered waves, spawning only unblocked tasks in parallel while respecting file ownership conflicts via pre-spawn manifest validation \n Implements Ralph Wiggum pattern: fresh worker

Purpose: Take a project from "code is ready" to "tagged and ready to push."

Purpose: Check the agent mail inbox for pending messages, HELP_REQUESTs, and inter-agent notifications.

Quick Ref: Autonomous epic execution. /swarm for each wave with runtime-native spawning. Output: closed issues + phase-2 handoff for /validation.

Purpose: Is this plan/spec good enough to implement?

YOU MUST EXECUTE THIS WORKFLOW. Do not just describe it.

Strategic planning for open source contributions.

Systematic exploration of upstream repositories before contributing.

Scaffold and audit documentation for open source projects.

Monitor the knowledge flywheel health.

Quick Ref: 4-phase investigation (Root Cause → Pattern → Hypothesis → Fix). Output: .agents/research/YYYY-MM-DD-bug-*.md

inject

Fork-based implementation for open source contributions with mandatory isolation check.

Trace knowledge artifact lineage to sources.

Systematic PR preparation that validates tests and generates high-quality PR bodies.