Skip to content

Expecho/Agent42 / .claude/skills/decide

decide

Verified

Record an architectural or technical decision as an ADR in .agentic/decisions/, checking for conflicts with existing decisions and handling supersession.

FileKindSize
SKILL.mdskill md1.5 KB
---
name: decide
description: Record an architectural or technical decision as an ADR in .agentic/decisions/, checking for conflicts with existing decisions and handling supersession.
user-invocable: true
argument-hint: "<short description of the decision>"
---

# Record a decision (ADR)

1. **Check prior art:** grep `.agentic/decisions/` for the area/tags. If an active
   ADR (no `superseded_by`) conflicts with the new decision, this is a supersession —
   confirm with the user before proceeding.
2. **Gather the substance.** From the conversation context or by asking: context,
   the decision, alternatives actually considered (with why each lost), consequences
   including accepted downsides. If alternatives are unknown, ask — an ADR without
   alternatives is a changelog entry, not a decision record.
3. **Write** `NNNN-<slug>.md` (next free number on `origin/main` — fetch first; a parallel branch may
   have taken it, and `AdrNumberingTests` fails on a duplicate) from `0000-adr-template.md`. Today's
   date, `status: accepted`, meaningful `tags`.
4. **On supersession:** set `supersedes: NNNN` in the new file AND edit the old ADR
   (`status: superseded`, `superseded_by: MMMM`). Same commit.
5. If the decision creates a non-obvious constraint agents will trip on, also add a
   gotcha entry (see `/gotcha`) pointing at the ADR as evidence context.

Do NOT log routine implementation choices — only decisions with real alternatives
and lasting consequences. When in doubt, ask the user whether it clears that bar.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 322kUpdated Claude CodeCodexVerifiedInstall