postmortem

A rollback follow-up command that records what happened after a released feature is turned off or the previous release is restored. It creates a blameless postmortem—a report focused on causes and improvements rather than blame—and reopens the work.

In plain words
What is it for?
Use it after a rollback to gather contract, deployment, guard-report, and Sentry data, write the postmortem, notify Slack, and create follow-up work.
Why use it?
It gives a failed release a documented ending and preserves the timeline, signals, contributing factors, and next steps instead of losing that information.

Command

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/iampawan/manifest/postmortem
Clone the repo
git clone --depth 1 https://github.com/iampawan/Manifest
Per session 61 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 610 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 $0.00061 $0.00610
Opus 5 $0.00030 $0.00305
Sonnet 5 $0.00012 $0.00122
Haiku 4.5 $0.00006 $0.00061

Measured 2d ago against content hash eed139841301, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

postmortem 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 2d ago.

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.

commands/postmortem.md · 56 lines

How it starts

The opening of the file, as written. The whole thing — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/postmortem

Closes the loop after a rollback — the one terminal state the pipeline otherwise had no ending for. Run it once you've actually reverted the feature (flag off / previous release redeployed); rollback-guard recommends the rollback, you perform it, then this records and learns from it.

Usage

/postmortem SC-005
/postmortem SC-005 "iOS crash on cold start at 10%"

What it does

  1. Confirms the feature was rolled back (asks if unclear — it won't write a postmortem for something still live).
  2. Reconstructs the timeline from the contract's cycle-time stamps, the guard-*.md reports, the deploy reports, and Sentry.
  3. Writes .manifest/contracts/<ID>.postmortem.md — blameless: what happened, the breached signal, contributing factors, what went well (detection caught it before full rollout), action items, re-ship plan.
  4. Stamps the contract: landed: rolled-back, rolledBackAt.
  5. Reopens the work — a follow-up tracking issue (never closes it as "done") or a route through /fix for small fixes.
  6. Posts the verdict to Slack (thread + a top-level alert) and DMs the owner.

Learns from it. If a contributing factor is a code-shaped bug class review could have caught, the postmortem proposes a candidate bug pattern in reference/bug-patterns.candidates.md (staging — not yet enforced). A maintainer reviews and, if it generalizes, promotes it into reference/BUG-PATTERNS.md, where code-review then checks it on every future diff — so the same class of bug can't ship twice. The human-accept gate and the warning-until-proven default keep a new pattern from blocking every PR. See the learning loop in BUG-PATTERNS.md.

Blameless by design

The postmortem describes systems, specs, and gaps — never a person as the cause. A rollback is information: detection worked and the blast radius was bounded. The feature still didn't land, and that's the thing to fix.

See also

  • /rollback-check <ID> / /canary <ID> — the rollout-window health + ramp
  • /contract verify <ID> — re-verify the new revision before re-promoting

Read the full file on GitHub · 56 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. 2d ago First seen · 56 lines · 61 tokens per session scan A eed139841301

Subscribe to this mod's changes

postmortem is a command published in the GitHub repository iampawan/Manifest (5 stars, last pushed 20d ago), licensed MIT. It adds 61 tokens to every session and 610 once invoked, about $0.0003 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.