mirror of https://github.com/azaion/detections-semantic.git synced 2026-04-23 02:56:38 +00:00

Files

T

Oleksandr Bezdieniezhnykh 8e2ecf50fd Initial commit

Made-with: Cursor

2026-03-26 00:20:30 +02:00

23 KiB

Raw Blame History

name, description, category, tags, disable-model-invocation

name

description

Autopilot Orchestrator

Auto-chaining execution engine that drives the full BUILD → SHIP workflow. Detects project state from _docs/, resumes from where work stopped, and flows through skills automatically. The user invokes /autopilot once — the engine handles sequencing, transitions, and re-entry.

Core Principles

Auto-chain: when a skill completes, immediately start the next one — no pause between skills
Only pause at decision points: BLOCKING gates inside sub-skills are the natural pause points; do not add artificial stops between steps
State from disk: all progress is persisted to _docs/_autopilot_state.md and cross-checked against _docs/ folder structure
Rich re-entry: on every invocation, read the state file for full context before continuing
Delegate, don't duplicate: read and execute each sub-skill's SKILL.md; never inline their logic here
Sound on pause: follow .cursor/rules/human-input-sound.mdc — play a notification sound before every pause that requires human input
Minimize interruptions: only ask the user when the decision genuinely cannot be resolved automatically
Jira MCP required: steps that create Jira artifacts (Plan Step 6, Decompose) must have authenticated Jira MCP — never skip or substitute with local files

Jira MCP Authentication

Several workflow steps create Jira artifacts (epics, tasks, links). The Jira MCP server must be authenticated before any step that writes to Jira.

Steps That Require Jira MCP

Step	Sub-Step	Jira Action
2 (Plan)	Step 6 — Jira Epics	Create epics for each component
3 (Decompose)	Step 1–3 — All tasks	Create Jira ticket per task, link to epic

Authentication Gate

Before entering Step 2 (Plan) or Step 3 (Decompose) for the first time, the autopilot must:

Call mcp_auth on the Jira MCP server
If authentication succeeds → proceed normally
If the user skips authentication → STOP. Present using Choose format:

══════════════════════════════════════
 BLOCKER: Jira MCP authentication required
══════════════════════════════════════
 A) Authenticate now (retry mcp_auth)
 B) Pause autopilot — resume after configuring Jira MCP
══════════════════════════════════════
 Note: Jira integration is mandatory. Plan and Decompose
 steps create epics and tasks that drive implementation.
 Local-only workarounds are not acceptable.
══════════════════════════════════════

Do NOT offer a "skip Jira" or "save locally" option. The workflow depends on Jira IDs for task referencing, dependency tracking, and implementation batching.

Re-Authentication

If Jira MCP was already authenticated in a previous invocation (verify by listing available Jira tools beyond mcp_auth), skip the auth gate.

User Interaction Protocol

Every time the autopilot or a sub-skill needs a user decision, use the Choose A / B / C / D format. This applies to:

State transitions where multiple valid next actions exist
Sub-skill BLOCKING gates that require user judgment
Any fork where the autopilot cannot confidently pick the right path
Trade-off decisions (tech choices, scope, risk acceptance)

When to Ask (MUST ask)

The next action is ambiguous (e.g., "another research round or proceed?")
The decision has irreversible consequences (e.g., architecture choices, skipping a step)
The user's intent or preference cannot be inferred from existing artifacts
A sub-skill's BLOCKING gate explicitly requires user confirmation
Multiple valid approaches exist with meaningfully different trade-offs

When NOT to Ask (auto-transition)

Only one logical next step exists (e.g., Problem complete → Research is the only option)
The transition is deterministic from the state (e.g., Plan complete → Decompose)
The decision is low-risk and reversible
Existing artifacts or prior decisions already imply the answer

Choice Format

Always present decisions in this format:

══════════════════════════════════════
 DECISION REQUIRED: [brief context]
══════════════════════════════════════
 A) [Option A — short description]
 B) [Option B — short description]
 C) [Option C — short description, if applicable]
 D) [Option D — short description, if applicable]
══════════════════════════════════════
 Recommendation: [A/B/C/D] — [one-line reason]
══════════════════════════════════════

Rules:

Always provide 2–4 concrete options (never open-ended questions)
Always include a recommendation with a brief justification
Keep option descriptions to one line each
If only 2 options make sense, use A/B only — do not pad with filler options
Play the notification sound (per human-input-sound.mdc) before presenting the choice
Record every user decision in the state file's Key Decisions section
After the user picks, proceed immediately — no follow-up confirmation unless the choice was destructive

State File: `_docs/_autopilot_state.md`

The autopilot persists its state to _docs/_autopilot_state.md. This file is the primary source of truth for re-entry. Folder scanning is the fallback when the state file doesn't exist.

Format

# Autopilot State

## Current Step
step: [0-5 or "done"]
name: [Problem / Research / Plan / Decompose / Implement / Deploy / Done]
status: [not_started / in_progress / completed]
sub_step: [optional — sub-skill internal step number + name if interrupted mid-step]

## Step ↔ SubStep Reference
| Step | Name       | Sub-Skill              | Internal SubSteps                        |
|------|------------|------------------------|------------------------------------------|
| 0    | Problem    | problem/SKILL.md       | Phase 1–4                                |
| 1    | Research   | research/SKILL.md      | Mode A: Phase 1–4 · Mode B: Step 0–8    |
| 2    | Plan       | plan/SKILL.md          | Step 1–6                                 |
| 3    | Decompose  | decompose/SKILL.md     | Step 1–4                                 |
| 4    | Implement  | implement/SKILL.md     | (batch-driven, no fixed sub-steps)       |
| 5    | Deploy     | deploy/SKILL.md        | Step 1–7                                 |

When updating `Current Step`, always write it as:
  step: N          ← autopilot step (0–5)
  sub_step: M      ← sub-skill's own internal step/phase number + name
Example:
  step: 2
  name: Plan
  status: in_progress
  sub_step: 4 — Architecture Review & Risk Assessment

## Completed Steps

| Step | Name | Completed | Key Outcome |
|------|------|-----------|-------------|
| 0 | Problem | [date] | [one-line summary] |
| 1 | Research | [date] | [N drafts, final approach summary] |
| 2 | Plan | [date] | [N components, architecture summary] |
| 3 | Decompose | [date] | [N tasks, total complexity points] |
| 4 | Implement | [date] | [N batches, pass/fail summary] |
| 5 | Deploy | [date] | [artifacts produced] |

## Key Decisions
- [decision 1: e.g. "Tech stack: Python + Rust for perf-critical, Postgres DB"]
- [decision 2: e.g. "6 research rounds, final draft: solution_draft06.md"]
- [decision N]

## Last Session
date: [date]
ended_at: Step [N] [Name] — SubStep [M] [sub-step name]
reason: [completed step / session boundary / user paused / context limit]
notes: [any context for next session, e.g. "User asked to revisit risk assessment"]

## Blockers
- [blocker 1, if any]
- [none]

State File Rules

Create the state file on the very first autopilot invocation (after state detection determines Step 0)
Update the state file after every step completion, every session boundary, and every BLOCKING gate confirmation
Read the state file as the first action on every invocation — before folder scanning
Cross-check: after reading the state file, verify against actual _docs/ folder contents. If they disagree (e.g., state file says Step 2 but _docs/02_plans/architecture.md already exists), trust the folder structure and update the state file to match
Never delete the state file. It accumulates history across the entire project lifecycle

Execution Entry Point

Every invocation of this skill follows the same sequence:

1. Read _docs/_autopilot_state.md (if exists)
2. Cross-check state file against _docs/ folder structure
3. Resolve current step (state file + folder scan)
4. Present Status Summary (from state file context)
5. Enter Execution Loop:
   a. Read and execute the current skill's SKILL.md
   b. When skill completes → update state file
   c. Re-detect next step
   d. If next skill is ready → auto-chain (go to 5a with next skill)
   e. If session boundary reached → update state file with session notes → suggest new conversation
   f. If all steps done → update state file → report completion

State Detection

Read _docs/_autopilot_state.md first. If it exists and is consistent with the folder structure, use the Current Step from the state file. If the state file doesn't exist or is inconsistent, fall back to folder scanning.

Folder Scan Rules (fallback)

Scan _docs/ to determine the current workflow position. Check rules in order — first match wins.

Detection Rules

Step 0 — Problem Gathering Condition: _docs/00_problem/ does not exist, OR any of these are missing/empty:

problem.md
restrictions.md
acceptance_criteria.md
input_data/ (must contain at least one file)

Action: Read and execute .cursor/skills/problem/SKILL.md

Step 1 — Research (Initial) Condition: _docs/00_problem/ is complete AND _docs/01_solution/ has no solution_draft*.md files

Action: Read and execute .cursor/skills/research/SKILL.md (will auto-detect Mode A)

Step 1b — Research Decision Condition: _docs/01_solution/ contains solution_draft*.md files AND _docs/01_solution/solution.md does not exist AND _docs/02_plans/architecture.md does not exist

Action: Present the current research state to the user:

How many solution drafts exist
Whether tech_stack.md and security_analysis.md exist
One-line summary from the latest draft

Then present using the Choose format:

══════════════════════════════════════
 DECISION REQUIRED: Research complete — next action?
══════════════════════════════════════
 A) Run another research round (Mode B assessment)
 B) Proceed to planning with current draft
══════════════════════════════════════
 Recommendation: [A or B] — [reason based on draft quality]
══════════════════════════════════════

If user picks A → Read and execute .cursor/skills/research/SKILL.md (will auto-detect Mode B)
If user picks B → auto-chain to Step 2 (Plan)

Step 2 — Plan Condition: _docs/01_solution/ has solution_draft*.md files AND _docs/02_plans/architecture.md does not exist

Action:

The plan skill's Prereq 2 will rename the latest draft to solution.md — this is handled by the plan skill itself
Read and execute .cursor/skills/plan/SKILL.md

If _docs/02_plans/ exists but is incomplete (has some artifacts but no FINAL_report.md), the plan skill's built-in resumability handles it.

Step 3 — Decompose Condition: _docs/02_plans/ contains architecture.md AND _docs/02_plans/components/ has at least one component AND _docs/02_tasks/ does not exist or has no task files (excluding _dependencies_table.md)

Action: Read and execute .cursor/skills/decompose/SKILL.md

If _docs/02_tasks/ has some task files already, the decompose skill's resumability handles it.

Step 4 — Implement Condition: _docs/02_tasks/ contains task files AND _dependencies_table.md exists AND _docs/03_implementation/FINAL_implementation_report.md does not exist

Action: Read and execute .cursor/skills/implement/SKILL.md

If _docs/03_implementation/ has batch reports, the implement skill detects completed tasks and continues.

Step 5 — Deploy Condition: _docs/03_implementation/FINAL_implementation_report.md exists AND _docs/04_deploy/ does not exist or is incomplete

Action: Read and execute .cursor/skills/deploy/SKILL.md

Done Condition: _docs/04_deploy/ contains all expected artifacts (containerization.md, ci_cd_pipeline.md, environment_strategy.md, observability.md, deployment_procedures.md)

Action: Report project completion with summary.

Status Summary

On every invocation, before executing any skill, present a status summary built from the state file (with folder scan fallback).

Format:

═══════════════════════════════════════════════════
 AUTOPILOT STATUS
═══════════════════════════════════════════════════
 Step 0  Problem      [DONE / IN PROGRESS / NOT STARTED]
 Step 1  Research     [DONE (N drafts) / IN PROGRESS / NOT STARTED]
 Step 2  Plan         [DONE / IN PROGRESS / NOT STARTED]
 Step 3  Decompose    [DONE (N tasks) / IN PROGRESS / NOT STARTED]
 Step 4  Implement    [DONE / IN PROGRESS (batch M of ~N) / NOT STARTED]
 Step 5  Deploy       [DONE / IN PROGRESS / NOT STARTED]
═══════════════════════════════════════════════════
 Current: Step N — Name
 SubStep: M — [sub-skill internal step name]
 Action:  [what will happen next]
═══════════════════════════════════════════════════

For re-entry (state file exists), also include:

Key decisions from the state file's Key Decisions section
Last session context from the Last Session section
Any blockers from the Blockers section

Auto-Chain Rules

After a skill completes, apply these rules:

Completed Step	Next Action
Problem Gathering	Auto-chain → Research (Mode A)
Research (any round)	Auto-chain → Research Decision (ask user: another round or proceed?)
Research Decision → proceed	Auto-chain → Plan
Plan	Auto-chain → Decompose
Decompose	Session boundary — suggest new conversation before Implement
Implement	Auto-chain → Deploy
Deploy	Report completion

Session Boundary: Decompose → Implement

After decompose completes, do not auto-chain to implement. Instead:

Update state file: mark Decompose as completed, set current step to 4 (Implement) with status not_started
Write Last Session section: reason: session boundary, notes: Decompose complete, implementation ready
Present a summary: number of tasks, estimated batches, total complexity points
Use Choose format:

══════════════════════════════════════
 DECISION REQUIRED: Decompose complete — start implementation?
══════════════════════════════════════
 A) Start a new conversation for implementation (recommended for context freshness)
 B) Continue implementation in this conversation
══════════════════════════════════════
 Recommendation: A — implementation is the longest phase, fresh context helps
══════════════════════════════════════

This is the only hard session boundary. All other transitions auto-chain.

Skill Delegation

For each step, the delegation pattern is:

Update state file: set step to the autopilot step number (0–5), status to in_progress, set sub_step to the sub-skill's current internal step/phase number and name
Announce: "Starting [Skill Name]..."
Read the skill file: .cursor/skills/[name]/SKILL.md
Execute the skill's workflow exactly as written, including:
- All BLOCKING gates (present to user, wait for confirmation)
- All self-verification checklists
- All save actions
- All escalation rules
- Update sub_step in the state file each time the sub-skill advances to a new internal step/phase
When the skill's workflow is fully complete:
- Update state file: mark step as completed, record date, write one-line key outcome
- Add any key decisions made during this step to the Key Decisions section
- Return to the auto-chain rules

Do NOT modify, skip, or abbreviate any part of the sub-skill's workflow. The autopilot is a sequencer, not an optimizer.

Re-Entry Protocol

When the user invokes /autopilot and work already exists:

Read _docs/_autopilot_state.md
Cross-check against _docs/ folder structure
Present Status Summary with context from state file (key decisions, last session, blockers)
If the detected step has a sub-skill with built-in resumability (plan, decompose, implement, deploy all do), the sub-skill handles mid-step recovery
Continue execution from detected state

Error Handling

All error situations that require user input MUST use the Choose A / B / C / D format.

Situation	Action
State detection is ambiguous (artifacts suggest two different steps)	Present findings and use Choose format with the candidate steps as options
Sub-skill fails or hits an unrecoverable blocker	Use Choose format: A) retry, B) skip with warning, C) abort and fix manually
User wants to skip a step	Use Choose format: A) skip (with dependency warning), B) execute the step
User wants to go back to a previous step	Use Choose format: A) re-run (with overwrite warning), B) stay on current step
User asks "where am I?" without wanting to continue	Show Status Summary only, do not start execution

Trigger Conditions

This skill activates when the user wants to:

Start a new project from scratch
Continue an in-progress project
Check project status
Let the AI guide them through the full workflow

Keywords: "autopilot", "auto", "start", "continue", "what's next", "where am I", "project status"

Differentiation:

User wants only research → use /research directly
User wants only planning → use /plan directly
User wants the full guided workflow → use /autopilot

Methodology Quick Reference

┌────────────────────────────────────────────────────────────────┐
│              Autopilot (Auto-Chain Orchestrator)                │
├────────────────────────────────────────────────────────────────┤
│ EVERY INVOCATION:                                              │
│   1. State Detection (scan _docs/)                             │
│   2. Status Summary (show progress)                            │
│   3. Execute current skill                                     │
│   4. Auto-chain to next skill (loop)                           │
│                                                                │
│ WORKFLOW:                                                       │
│   Step 0  Problem    → .cursor/skills/problem/SKILL.md         │
│     ↓ auto-chain                                               │
│   Step 1  Research   → .cursor/skills/research/SKILL.md        │
│     ↓ auto-chain (ask: another round?)                         │
│   Step 2  Plan       → .cursor/skills/plan/SKILL.md            │
│     ↓ auto-chain                                               │
│   Step 3  Decompose  → .cursor/skills/decompose/SKILL.md       │
│     ↓ SESSION BOUNDARY (suggest new conversation)              │
│   Step 4  Implement  → .cursor/skills/implement/SKILL.md       │
│     ↓ auto-chain                                               │
│   Step 5  Deploy     → .cursor/skills/deploy/SKILL.md          │
│     ↓                                                          │
│   DONE                                                         │
│                                                                │
│ STATE FILE: _docs/_autopilot_state.md                          │
│ FALLBACK: _docs/ folder structure scan                         │
│ PAUSE POINTS: sub-skill BLOCKING gates only                    │
│ SESSION BREAK: after Decompose (before Implement)              │
│ USER INPUT: Choose A/B/C/D format at genuine decisions only    │
│ AUTO-TRANSITION: when path is unambiguous, don't ask            │
├────────────────────────────────────────────────────────────────┤
│ Principles: Auto-chain · State to file · Rich re-entry         │
│             Delegate don't duplicate · Pause at decisions only  │
│             Minimize interruptions · Choose format for decisions │
└────────────────────────────────────────────────────────────────┘

23 KiB Raw Blame History Unescape Escape