agile-v/agile_v_skills/agile-v-adr

Claude Code and agent skill by agile-v/agile_v_skills

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Authoring, approval, immutability, and supersession of Architecture Decision Records (ADRs) in the Agile V lifecycle. Load when recording a significant, long-lived architectural, platform, tooling, or security decision.

Install

npx vskill@latest install agile-v/agile_v_skills/agile-v-adr

SKILL.md

Instructions

An Architecture Decision Record (ADR) captures a significant, long-lived technical decision. It is a distinct artifact from the append-only DECISION_LOG.md: an ADR has a fixed field set, is immutable once written, and is changed only by supersession. Agile V reserves the ADR-XXXX ID; this skill backs it with an artifact and procedure.

Requires agile-v-core loaded first. An ADR does not replace a REQ, risk control, specification, or change record.

When to write an ADR

Use an ADRUse DECISION_LOG.md
Architecturally significant, long-lived choice (platform, framework, tool, architecture pattern, infrastructure, data, security posture)Routine, reversible, day-to-day implementation decisions
A choice future maintainers must not silently reverseSession/context notes and rationale
A decision affecting regulated functionality, data integrity, interfaces, validation, or lifecycle risk—

If in doubt, an ADR is the more durable, auditable choice.

ADR fields (append-only, immutable once written)

FieldValue / allowed values
IDADR-XXXX — unique, never reused
TypePlatform \Framework \Tool \Architecture \Infrastructure \Data \Security
DateDate of decision
Statusproposed \approved \active \deprecated \superseded
SupersedesReference to the ADR this one replaces (if any)
Review DateScheduled re-evaluation date
ContextForces / problem driving the decision
Options ConsideredAlternatives evaluated
DecisionThe choice made
RationaleWhy this option over the others
ConsequencesTrade-offs, follow-on constraints

Procedure

  1. Propose — create the ADR with Status: proposed; Context, Options

Considered, Decision, Rationale, Consequences are mandatory.

  1. Approve — the responsible human approver reviews (Human Gate etiquette,

agile-v-core Directive 5). On approval, proposed → approved → active. Record approval evidence if the ADR gates a baseline or release.

  1. Link — relate affected requirements, risks, specifications, and change

records via typed relationships (agile-v-core Directive 2).

Immutability & supersession

  • An active ADR is never edited in place.
  • To reverse or amend, write a new ADR with Supersedes: ADR-YYYY, and set

the old one to Status: superseded (the only permitted status edit on a written ADR).

  • On Review Date, re-evaluate: keep active, or supersede.

Storage & traceability

  • One ADR per record under .agile-v/adr/ (or a dedicated ADR_LOG.md),

separate from the append-only DECISION_LOG.md.

  • The DECISION_LOG.md may note an ADR promotion with a pointer to ADR-XXXX,

but the ADR is the governed artifact.

  • When an external ALM/requirements tool is bound via an extension, ADRs map to

a dedicated item type in that tool — not folded into a decision-log item.

Halt conditions

  • ADR proposed without Context, Options, Decision, Rationale, or Consequences.
  • In-place edit of an approved/active ADR (use supersession instead).
  • Reuse of a retired ADR-XXXX ID.

Security

  • Security scanPassed, 95 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
54
Repository forks
10
7-day trend score
20

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/agile-v/agile_v_skills/agile-v-adr/badge)](https://verified-skill.com/skills/agile-v/agile_v_skills/agile-v-adr)

Something wrong with this skill or its page? Tell us.