Capturing Decisions
Skill0 starsChanged 10 months agoApache-2.0View on GitHub
Create and maintain MADR-format Architectural Decision Records in Markdown under `docs/decisions`. Use when the user wants to document important decisions.
In this repository5 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .claude/CLAUDE.mdCLAUDE.md
- generating-pikchrSkill
- markdown-authoringSkill
- using-gitSkill
The file
We split this skill into the rules it gives an agent
9 rules. About 0 are common, 5 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 17
All ADRs live in:
- Very specific: line 26
If docs/decisions/ does not exist yet, create it.
- Distinctive: line 30
Use this copy of the MADR 4.0.0 long-form template
- Distinctive: line 32
Create new ADRs by copying the template and replacing placeholders.
- Very specific: line 62
Default status to proposed unless the change set includes implementation and agreement to accept.
- Distinctive: line 64
Add the ADR in numeric order.
- Very specific: line 65
Ensure the emoji matches the ADR's status.
- Very specific: line 66
If an ADR is superseded, keep the old ADR file, set its status to superseded by ADR-NNNN, and update the index row emoji to 🔁.
- Distinctive: line 71
Keep ADRs concise, but include enough context that a new reader can understand why the decision was made.
≈ “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.
Files
MADR ADRs
This skill creates (and keeps tidy) Architectural Decision Records (ADRs) using MADR (Markdown Architectural Decision Records) v4.0.0.
A helpful bar for what "counts" as an architectural decision is Martin Fowler's: "a decision you wish you could get right early." ("Who needs an architect?", IEEE Software, 2003)
Primary reference: Markdown Architectural Decision Records
Repository layout and naming
All ADRs live in:
docs/decisions/NNNN-title-with-dashes.md
Where:
NNNNis a zero-padded 4-digit sequence number (0001,0002, …)title-with-dashesis a lowercase slug (letters/digits/hyphens)
If docs/decisions/ does not exist yet, create it.
Template
Use this copy of the MADR 4.0.0 long-form template
Create new ADRs by copying the template and replacing placeholders. Optional sections may be removed (the template marks them clearly).
Required metadata
Each ADR must include YAML front matter at the top with:
status: one ofproposed,accepted,rejected,deprecated, orsuperseded by ADR-NNNNdate:YYYY-MM-DD(update when the ADR is materially changed)
Status emoji for the index
Maintain an index at docs/decisions/README.md that lists all ADRs with:
- status emoji
- ADR title (matching the H1 of the ADR), as a link to the full ADR
- date last updated
Use this mapping:
- 🟡 proposed
- ✅ accepted
- ❌ rejected
- ⚠️ deprecated
- 🔁 superseded
Process
- Pick the next number by scanning existing ADR filenames in
docs/decisions/and incrementing the highestNNNN. Start at0001if none exist. - Slugify the title into
title-with-dashes(lowercase, hyphens, no punctuation). - Create the ADR from
ADR_TEMPLATE.md.- Default
statustoproposedunless the change set includes implementation and agreement to accept.
- Default
- Update
docs/decisions/README.md:- Add the ADR in numeric order.
- Ensure the emoji matches the ADR's
status.
- If an ADR is superseded, keep the old ADR file, set its status to
superseded by ADR-NNNN, and update the index row emoji to 🔁.
Output expectations
- ADR markdown should be clean and readable in GitHub rendering.
- Keep ADRs concise, but include enough context that a new reader can understand why the decision was made.
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
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.

