post-mortem

A command that reviews an incident after it happens by comparing a context-rich account with a timeline-only account. It is designed to reveal when a team settles too quickly on blaming a person or familiar failure.

In plain words
What is it for?
Second-pass post-mortems: identifying the immediate trigger, listing candidate root causes with confidence levels, checking blame anchoring, and finding unanswered questions.
Why use it?
It helps separate evidence from assumptions and exposes possible technical, process, and organisational causes that a first review may miss.

Command for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/heznpc/ploidy/post-mortem
Clone the repo
git clone --depth 1 https://github.com/heznpc/PLOIDY

Made for: Claude Code.

Per session 43 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 944 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00043 $0.00944
Opus 5 $0.00022 $0.00472
Sonnet 5 $0.00009 $0.00189
Haiku 4.5 $0.00004 $0.00094

Measured yesterday against content hash a2d6c33ef27b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

post-mortem scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/post-mortem.md · 108 lines

How it starts

The opening of the file, as written. The whole thing — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.

The user wants a second pass on an incident analysis:

$ARGUMENTS

You hold the context: who was on-call, what the team already blamed, how similar incidents played out, which systems had prior issues. That context is how blame-anchoring creeps in — once "ah, it was the retry storm again" becomes the narrative, reviewers stop looking. Run the four-step pass without asking.

1 · Write the Deep-context take

If $ARGUMENTS is a path (ends in .md / contains /), read the file first. Otherwise treat it as the incident summary.

In ~300 words:

  1. Proximate cause — the one-step-back trigger, stated plainly.
  2. Root causes (candidates) — list every plausible contributing factor. For each: technical / process / organisational.
  3. Blame anchoring check — list who or what in this incident the team is already converging on as responsible. For each, state whether the convergence is (a) backed by the timeline or (b) convenient because the target has a track record.
  4. Unasked questions — three questions the post-mortem has not asked but probably should.

Tag every candidate cause HIGH / MEDIUM / LOW confidence.

2 · Spawn a Fresh sub-agent

Use the Agent tool (subagent_type="general-purpose"). It must see only the timeline of events, no names of teams, no history of prior incidents.

Sanitise the input: strip team / person names, prior-incident references, and any "as usual / again / typical" phrases. Replace names with roles ("on-call engineer", "service owner"). Keep technical facts intact.

Prompt the subagent with:

You are reading the timeline of an incident in a system you have never seen. Only the sanitised timeline is available:

Answer in under 250 words:

  1. What is the most likely proximate cause, stated as one sentence?
  2. List every plausible root cause you can infer from the timeline alone. For each, say what the timeline would need to show to confirm or refute it.
  3. What information is missing from the timeline that, if present, would change your assessment?
  4. What recurring failure patterns in generic distributed systems does this timeline match (retry storm, thundering herd, partial deployment, clock skew, quota exhaustion, …)?

Do not ask for context. Do not guess at team / org details.

Read the full file on GitHub · 108 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 108 lines · 43 tokens per session scan A a2d6c33ef27b

Subscribe to this mod's changes

post-mortem is a command published in the GitHub repository heznpc/PLOIDY (0 stars, last pushed 5d ago), licensed MIT. It adds 43 tokens to every session and 944 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.