pre-mortem

What this skill does

# Pre-Mortem Skill

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

> **Mandatory for 3+ issue epics.** Pre-mortem is enforced by hook when `/crank` is invoked on epics with 3+ child issues. 6/6 consecutive positive ROI. Bypass: `--skip-pre-mortem` flag or `AGENTOPS_SKIP_PRE_MORTEM_GATE=1`.

## Loop position

Pre-flight check between moves **3 (slice plan)** and **4 (TDD per slice)** of the [operating loop](../../docs/architecture/operating-loop.md). Consumes the [slice validation plan](../../docs/templates/slice-validation.md); produces a PASS/WARN/FAIL verdict on the plan AND on the wave-validity rows (distinct write scopes, no shared migration/contract/CLI surface, owner per slice, discard path per slice). A wave can only be claimed parallel if pre-mortem confirms every conflict-free row. FAIL on wave-validity → run slices sequential or send the plan back to `/plan` for re-slicing.

Run `/council validate` on a plan or spec to get multi-model judgment before committing to implementation.

---

## Quick Start

```bash
/pre-mortem                                         # validates most recent plan (inline, no spawning)
/pre-mortem path/to/PLAN.md                         # validates specific plan (inline)
/pre-mortem --deep path/to/SPEC.md                  # 4 judges (thorough review, spawns agents)
/pre-mortem --mixed path/to/PLAN.md                 # cross-vendor (Claude + Codex)
/pre-mortem --preset=architecture path/to/PLAN.md   # architecture-focused review
/pre-mortem --explorers=3 path/to/SPEC.md           # deep investigation of plan
/pre-mortem --debate path/to/PLAN.md                # two-round adversarial review
```

---

## Execution Steps

### Step 0: Bead-Input Pre-Flight (Mandatory)

When the input to `/pre-mortem` is a bead ID (matches pattern `[a-z]{2,6}-[0-9a-z.]+`) AND complexity is "full" OR the bead is older than 7 days OR the bead description was filed by a prior session, automatically run `ao beads verify <bead-id>` as the very first action. If verify reports any STALE citations, present them to the user and ask for scope re-validation before proceeding. This implements the shared stale-scope validation rule.

### Step 1: Find the Plan/Spec

**If path provided:** Use it directly.

**If no path:** Find most recent plan:
```bash
ls -lt .agents/plans/ 2>/dev/null | head -3
ls -lt .agents/specs/ 2>/dev/null | head -3
```

Use the most recent file. If nothing found, ask user.

### Step 1.4: Retrieve Prior Learnings & Compiled Prevention (Mandatory)

Run `ao lookup` for the plan's domain, then load compiled checks from `.agents/pre-mortem-checks/*.md` (fall back to `.agents/findings/registry.jsonl`). Include matched entries in the council packet as `known_risks` and record `ao metrics cite` influence. Full contract (fail-open rules, section-evidence handling, ranking heuristics, citation lifecycle) in [references/compiled-prevention.md](references/compiled-prevention.md). This file also contains Step 1a (flywheel search, skipped under `--quick`) and Step 1b (PRODUCT.md auto-include).

Fail-open reader behavior is mandatory: missing or empty compiled prevention inputs skip silently; malformed line -> warn and ignore that line; unreadable file -> warn once and continue without findings.

### Step 1a: Flywheel Search (Skip if --quick)

Run the flywheel search from [references/compiled-prevention.md](references/compiled-prevention.md) unless `--quick` is active.

### Step 1b: PRODUCT.md Context (Skip if --quick)

When `PRODUCT.md` exists and full council mode is active, add one product judge: 3 judges total (2 plan-review + 1 product).

### Step 1.5: Fast Path (--quick mode)

**By default, pre-mortem runs inline (`--quick`)** — single-agent structured review, no spawning. This catches real implementation issues at ~10% of full council cost (proven in ag-nsx: 3 actionable bugs found inline that would have caused runtime failures).

In `--quick` mode, skip Steps 1a and 1b as standalone pre-processing phases. If `PRODUCT.md` exists, Step 1b's product context is still loaded inline during the quick review. `--deep`, `--mixed`, `--debate`, and `--explorers` add the dedicated product perspective and wider council fan-out.

To escalate to full multi-judge council, use `--deep` (4 judges) or `--mixed` (cross-vendor).

### Step 1.6: Scope Mode Selection

Determine review posture — EXPANSION, HOLD SCOPE, or REDUCTION — and commit `scope_mode: <expansion|hold|reduction>` in the council packet. Auto-detection rules and mode-specific judge prompts are in [references/scope-mode.md](references/scope-mode.md).

### Step 1.7: Load Council FAIL Patterns (Mandatory)

Read [references/council-fail-patterns.md](references/council-fail-patterns.md) for the top 8 council FAIL patterns to check against. These patterns are derived from 124 analyzed FAIL verdicts across 946 council sessions. They apply to both `--quick` and `--deep` modes.

### Step 2: Run Council Validation

**Default (inline, no spawning):**
```
/council --quick validate <plan-path>
```
Single-agent structured review. Catches real implementation issues at ~10% of full council cost. Sufficient for most plans (proven across 6+ epics).

Default (2 judges with plan-review perspectives) applies when you intentionally run non-quick council mode.

**With --deep (4 judges with plan-review perspectives):**
```
/council --deep --preset=plan-review validate <plan-path>
```
Spawns 4 judges:
- `missing-requirements`: What's not in the spec that should be? What questions haven't been asked?
- `feasibility`: What's technically hard or impossible here? What will take 3x longer than estimated?
- `scope`: What's unnecessary? What's missing? Where will scope creep?
- `spec-completeness`: Are boundaries defined? Do conformance checks cover all acceptance criteria? Is the plan mechanically verifiable?

Use `--deep` for high-stakes plans (migrations, security, multi-service, 7+ issues).

**With --mixed (cross-vendor):**
```
/council --mixed --preset=plan-review validate <plan-path>
```
3 Claude + 3 Codex agents for cross-vendor plan validation with plan-review perspectives.

**With explicit preset override:**
```
/pre-mortem --preset=architecture path/to/PLAN.md
```
Explicit `--preset` overrides the automatic plan-review preset. Uses architecture-focused personas instead.

**With explorers:**
```
/council --deep --preset=plan-review --explorers=3 validate <plan-path>
```
Each judge spawns 3 explorers to investigate aspects of the plan's feasibility against the codebase. Useful for complex migration or refactoring plans.

**With debate mode:**
```
/pre-mortem --debate
```
Enables adversarial two-round review for plan validation. Use for high-stakes plans where multiple valid approaches exist. See `/council` docs for full --debate details.

### Steps 2.4–2.8: Mandatory Council Checks

Five mandatory checks run during council validation — temporal interrogation, error-&-rescue map, council FAIL pattern check, test pyramid coverage, and input validation for enum-like fields. Each has auto-trigger conditions and judge-prompt snippets. Full step text and check tables in [references/mandatory-checks.md](references/mandatory-checks.md).

When a plan introduces a regex, grep, glob, or similar scope predicate, also apply [references/scope-predicate-positive-negative-cases.md](references/scope-predicate-positive-negative-cases.md): require positive and negative examples before approval.

### Step 2.9: No-self-grading invariant (author ≠ validator)

The pre-mortem verdict must NOT be graded by the plan's own author. A verdict produced by the authoring context is autocorrelated — the same assumptions that shaped the plan pass it. This is the no-self-grading invariant (`ag-lmdx.4`): the independent-trust-domain check on the plan-acceptance verdict.

**Rule:** the judge context MUST be distinct from the author context. Validation MAY run inside the authoring session, but the judge MUST be a **blind sub-agent** — a f