agentleFS
Sign inStart free
ovargasovargas/virtual-team/skills/adr-convention/SKILL.md

adr-convention

Skill14 starsChanged 8 months agoMITView on GitHub

Use when creating, reviewing, or referencing architectural decision records — defines the ADR format, three-gate creation threshold, and local/hub flavors

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Only create an ADR when ALL THREE gates are true:

    line 10
  2. Distinctive:

    If any gate fails, the decision is normal engineering — document it in the spec, plan, or code comment instead.

    line 16
  3. Very specific:

    Default type is contract unless the agreement is a convention or infrastructure decision

    line 57

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

ADR Convention

Three-Gate Threshold

Only create an ADR when ALL THREE gates are true:

  1. Hard to reverse — changing this decision later requires significant rework
  2. Surprising without context — a future reader would wonder "why did they do it this way?"
  3. Real trade-off — genuine alternatives existed with different pros/cons

If any gate fails, the decision is normal engineering — document it in the spec, plan, or code comment instead.

Format

File: docs/decisions/YYYY-MM-DD-<kebab-slug>.md

---
id: ADR-NNN        # Sequential. Scan docs/decisions/*.md for highest existing ID, increment.
date: YYYY-MM-DD
status: accepted   # accepted | superseded
type: technical    # technical | contract | convention | infrastructure | data
epic: EPIC-NNN     # Optional. Link to parent epic if applicable.
repos: [repo-a]    # Hub flavor only. Omit for local decisions.
---

Four required body sections:

# [Decision Title]

## Context
[Why this decision was needed. What problem or constraint triggered it.]

## Decision
[What was decided. Concrete, actionable statements.]

## Alternatives Considered
[What else was evaluated and why it was rejected. Brief — one line per alternative is fine.]

## Consequences
[What follows from this decision. Cross-reference enforcement docs if they exist.]

Hub Variant

Hub agreements (cross-repo decisions) add:

  • repos: frontmatter field (required) — which repos are affected
  • ## Affected Repos section with per-repo details
  • ### Details sub-section under Decision for repo-specific implementation notes
  • Default type is contract unless the agreement is a convention or infrastructure decision

Pipeline Integration

CommandRoleBehavior
planActiveThree-gate check after plan review; prompt to create ADR or flag for review
featureLightPrompt on YAGNI overrides and non-obvious approach decisions
tech-reviewPassiveRead existing ADRs; tag findings as ADR candidates; never auto-create
startReferencePoint to this skill for local ADR format
epicReferencePoint to this skill for hub ADR format

ID Generation

Scan all docs/decisions/*.md files for id: ADR-NNN in frontmatter. Take the highest number, increment by one. If no ADR files exist, start at ADR-001.

Related

  • ADR-001 defines the skill size budget: docs/decisions/2026-04-29-skill-size-budget.md
  • ADR-002 defines the domain glossary: docs/decisions/2026-04-29-domain-glossary-convention.md
  • ADR-003 defines token mode scope: docs/decisions/2026-04-29-token-mode-scope.md
  • skills/skill-authoring/SKILL.md enforces the 100-line budget from ADR-001

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free