Files
boshu2__agentops/docs/workflows
Bo b84dcf3ca2 docs(hookless): close cold-start orientation drift (soc-qh2g1 #cold-start-orientation-coherence) (#795)
## Summary

Close the hookless cold-start orientation gap for `soc-qh2g1` by making
the live startup docs and skills point at explicit `ao session
bootstrap` plus pulled context (`ao inject` / `ao lookup`) instead of
hidden SessionStart behavior.

## Changes

- Reframes live docs that still implied automatic startup/handoff
injection to the hookless explicit-command path.
- Updates source and Codex skill mirrors for `quickstart`, `inject`,
`using-agentops`, `recover`, `bug-hunt`, and `session-bootstrap`.
- Expands the cold-start and docs drift gates so `SessionStart hook`,
`startup hooks`, and `auto-inject ... hook` promises fail unless
explicitly hedged as opt-in, removed, or historical.
- Refreshes scoped Codex skill hashes for the touched mirrors.

## Test plan

- [x] `bash scripts/check-hookless-cold-start.sh`
- [x] `bash scripts/check-doc-hooks-drift.sh`
- [x] `bats tests/scripts/check-hookless-cold-start.bats
tests/scripts/check-doc-hooks-drift.bats`
- [x] `shellcheck scripts/check-hookless-cold-start.sh
scripts/check-doc-hooks-drift.sh`
- [x] `bash scripts/regen-codex-hashes.sh --check --only
session-bootstrap,quickstart,inject,using-agentops,recover,bug-hunt`
- [x] `bash scripts/validate-codex-generated-artifacts.sh`
- [x] `bash scripts/validate-codex-lifecycle-guards.sh`
- [x] `bash scripts/validate-agents-split.sh`
- [x] `git diff --check`
- [x] `bash scripts/validate-skill-runtime-parity.sh`
- [x] `bash scripts/lint-codex-native.sh`

Closes-scenario: soc-qh2g1#cold-start-orientation-coherence
Bounded-context: BC1-Corpus
Evidence: scripts/check-hookless-cold-start.sh
2026-06-06 11:23:56 -04:00
..

Workflow Guide

Which workflow should I use? Find the right one in seconds.

Golden Paths

Use these as the default user journeys before reaching for the full decision tree.

Situation Start with Continue with
First repo setup ao quick-start /quickstart, then /rpi "a small goal"
Small, obvious change Quick Fix /implement, then /vibe --quick recent
New feature or broad change Complete Cycle /rpi "goal" or /plan -> /crank <epic-id>
Existing tracked backlog ao reconcile --json, then bd ready /implement <issue-id> or /crank <epic-id>
Review-only checkpoint /council validate this PR /vibe recent before merge
Terminal-native operation ao factory start --goal "goal" ao rpi phased "goal" and ao rpi status

Decision Tree

I want to... Use this workflow Complexity Key Skills
Fix a simple bug or make a quick change Quick Fix Low /implement
Investigate and fix a hard bug Debug Cycle Medium /bug-hunt, /implement, /retro
Build a new feature end-to-end Complete Cycle High /research, /plan, /crank, /vibe
Deploy infrastructure changes Infrastructure Deployment High /plan, /implement, /vibe
Work across multiple domains at once Multi-Domain High /plan, /swarm
Validate research assumptions before planning Assumption Validation Medium /research
Understand a new codebase or extract patterns Knowledge Synthesis Medium /research, /knowledge-activation
Run a retrospective after completing work Post-Work Retro Low /retro, /post-mortem
Improve system quality over time Continuous Improvement Medium /retro, /vibe
Manage a session from start to finish Session Lifecycle Low (natural language)
Coordinate N autonomous workers without a leader Meta-Observer Pattern High /swarm

Workflows

Quick Fix

When to use: Simple, low-risk changes to 1-2 files with an obvious solution. Complexity: Low (10-30 min, single session) Key skills: /implement

Debug Cycle

When to use: Production incidents, intermittent failures, performance issues, or integration bugs with unknown root cause. Complexity: Medium-High (1-3 hours, 1-3 sessions) Key skills: /bug-hunt, /implement, /retro

Complete Cycle

When to use: New features, complex multi-file changes, or architectural work that needs full Research-Plan-Implement-Validate-Learn phases. Complexity: High (2-4 hours, 2-4 sessions) Key skills: /research, /plan, /crank, /vibe, /retro

Infrastructure Deployment

When to use: Infrastructure deployments that need validation gates, reality checks, and tracer bullets at every phase. Complexity: High (multi-session) Key skills: /plan, /implement, /vibe

Multi-Domain

When to use: Work spanning multiple domains simultaneously (backend + frontend + infra, technical + personal tracking). Complexity: High (multi-session) Key skills: /plan, /swarm

Assumption Validation

When to use: Before planning, to verify that research findings match reality (APIs exist, images available, operators behave as expected). Complexity: Medium Key skills: /research

Knowledge Synthesis

When to use: Onboarding to a new codebase, creating comprehensive docs, extracting patterns from multiple implementations. Complexity: Medium (30-60 min, single session) Key skills: /research, /knowledge-activation

Post-Work Retro

When to use: After completing significant work, to capture learnings and identify improvements. Complexity: Low Key skills: /retro, /post-mortem

Continuous Improvement

When to use: Periodic reviews (weekly/monthly), after milestones, technical debt reduction, or pattern refinement. Complexity: Medium (ongoing, periodic) Key skills: /retro, /vibe

Session Lifecycle

When to use: Managing any working session from start to finish -- starting work, resuming context, wrapping up. Complexity: Low (use natural language) Key skills: Just talk naturally; commands are optional.

Meta-Observer Pattern

When to use: Coordinating N autonomous workers through shared memory (stigmergy) without central orchestration. Complexity: High (advanced, multi-session) Key skills: /swarm Deep dive: meta-observer/ (pattern guide, examples, showcase)

Flowchart

Start here
    |
    v
Is it a simple, obvious fix? ----YES----> Quick Fix
    |
    NO
    |
    v
Is something broken? ----YES----> Debug Cycle
    |
    NO
    |
    v
Am I deploying infrastructure? ----YES----> Infrastructure Deployment
    |
    NO
    |
    v
Does it span multiple domains? ----YES----> Multi-Domain
    |
    NO
    |
    v
Am I building something new? ----YES----> Complete Cycle
    |
    NO
    |
    v
Am I trying to understand code/patterns? ----YES----> Knowledge Synthesis
    |
    NO
    |
    v
Am I wrapping up or improving? ----YES----> Post-Work Retro / Continuous Improvement