adr-discovery
ADR Discovery walks you through pre-flight context gathering before drafting an architecture decision record. It scans your repo for signals, confirms your domain and components one at a time, maps relationships, and classifies existing ADRs—outputting a human-verified discovery brief and an open-questions register. Hand off to adr-drafting once five required confirmations are locked.
ADR Discovery gathers and confirms architectural context before you draft an ADR, ensuring only verified facts enter your decision record.
AI-generated summary based on this skill's SKILL.md
Decision gist · record as of 2026-06-18
ADR Discovery gathers and confirms architectural context before you draft an ADR, ensuring only verified facts enter your decision record. ADR Discovery walks you through pre-flight context gathering before drafting an architecture decision record. It scans your repo for signals, confirms your domain and components one at a time, maps relationships, and classifies existing ADRs—outputting a human-verified discovery brief and an open-questions register. Hand off to adr-drafting once five required confirmations are locked.
Use it when
- ADR Discovery scans your repository for architectural signals.
- An ADR discovery brief is a human-verified summary of your architectural context—domain, confirmed components, relationships.
Verify before relying
Read SKILL.md below before installing (3 files). Open directory: indexed for reading, not audited.
Install
JosiahSiegel/claude-plugin-marketplace/adr-discovery · repository language: Shell
Open directory. Skills are indexed for reading, not audited. Review a skill's body before installing it.
Frequently asked questions
AI-generated answers based on this skill's SKILL.md and metadata
What information do I need before drafting an ADR?
ADR Discovery helps you gather and confirm all essential context before you write. Start by scanning your repo for signals about your domain, system components, and existing decisions. ADR Discovery then walks you through confirming each component one at a time, mapping relationships between them, and classifying any ADRs already in your decision log. Once you've locked five required confirmations, you'll have a human-verified discovery brief and an open-questions register ready to hand off to adr-drafting.
How does ADR Discovery map system components and relationships?
ADR Discovery scans your repository for architectural signals, then prompts you to confirm your domain and each component individually rather than guessing. As you confirm components, ADR Discovery helps you map how they relate to each other and to your overall system. It also classifies any existing architecture decision records so you understand what decisions have already been made and where gaps or unknowns remain in your architecture.
What is an ADR discovery brief and when should I use it?
An ADR discovery brief is a human-verified summary of your architectural context—domain, confirmed components, relationships, existing decisions, and constraints—that ADR Discovery produces once your five required confirmations are locked. Use it as your handoff document to adr-drafting. It contains only facts you've explicitly confirmed, not guesses or hallucinations, so your ADR drafting process starts from solid ground.
How does ADR Discovery help avoid hallucination in architecture docs?
ADR Discovery requires you to confirm architectural facts one at a time rather than accepting bulk assumptions. It builds an open-questions register for anything unconfirmed or unclear, and outputs only human-verified facts in your discovery brief. This deliberate, step-by-step confirmation process keeps your architecture documentation grounded in reality and flags unknowns explicitly so decision-makers know what still needs validation.
What is the ADR discovery workflow and how do I start?
ADR Discovery's workflow begins with a repo scan for architectural signals, then guides you through confirming your domain, system components, component relationships, existing decisions, and decision-makers or constraints. Each confirmation is locked before moving forward. Once all five required confirmations are complete, ADR Discovery outputs your discovery brief and open-questions register, ready to hand off to adr-drafting for the actual ADR draft.
Can ADR Discovery help me set up a decision log or ADR intake process?
Yes. ADR Discovery classifies existing ADRs in your repo and helps you understand your current decision-logging state. It identifies gaps in your decision log and surfaces unknowns through its open-questions register. This context helps you establish or refine your ADR intake process and decision log structure before you start drafting new records, ensuring consistency and completeness from the start.
SKILL.md
Rendered from the published skill. Quoted content, verbatim.
adr-discovery
Pre-flight context gathering for an ADR. The output is a docs/architecture/discovery-brief.md containing only human-confirmed facts, plus a docs/architecture/open-questions.md register of unknowns. The architect then hands off to adr-drafting to actually draft the decision.
Core rule: zero hallucination
No fact enters the discovery brief until the architect says yes. Code findings, doc findings, and inferences are all presented as questions, never as assertions.
| Wrong | Right |
|---|---|
| "Your system uses Postgres and Redis." | "I see pg and ioredis in package.json. Is |
(truncated - see the full file via the links below)
File tree — 3 files
plugins/doc-master/skills/adr-discovery/SKILL.md
plugins/doc-master/skills/adr-discovery/references/live-diagram-mode.md
plugins/doc-master/skills/adr-discovery/references/open-questions-register.md
Let your AI agent find skills like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 56,283 agent skills by what they can do, searchable in plain language.
wish › “Gather and confirm architectural context before drafting an ADR”
Give your agent the search over MCP, or paste the wish link into any chat. No install? Search from any chat →
Related skills
Guides you through seven-phase ADR drafting via focused dialogue, asking one question per turn and pressure-testing your reasoning before self-critique. Enforces specificity on components, deciders, and architectural characteristics, then routes to discovery if foundations are shallow.
c4-model produces LikeC4 specifications that enforce Simon Brown's canonical C4 model, generating context and container views alongside architecture decisions. It scaffolds the necessary DSL files and validates against a strict checklist, refusing component views, dynamic diagrams, custom element kinds, and nested systems that fall outside canonical C4 scope.
adr-critique systematically reviews existing or inherited ADRs to identify violations of shared rules, gaps in business-to-architecture reasoning, and conflicts with neighboring decisions or LikeC4 models. Each issue surfaces as a single flag requiring architect sign-off before moving to the next, ensuring quality without bulk rewrites.
adr-backfill lets you record decisions that shipped long ago but were never written up as ADRs—typically surfaced by doc audits or code archaeology. It enforces stricter honesty than forward drafting by requiring a visible backfill marker so future readers know the record was reconstructed, not contemporaneous. The skill demands two independent evidence sources and refuses to draft when confidence in the reasoning is too low.
Research Lint performs health checks on your research directory, identifying orphaned sources, missing entity hubs, broken wikilinks, stale claims, and contradictions. The skill runs seven checks—four fast mechanical scans plus optional LLM-powered analysis—and auto-applies safe edits like appending new contradictions and open questions while flagging issues that need your judgment.
Senior Architect guides system design through structured phases: requirements analysis, architecture evaluation, and decision documentation. It produces Architecture Decision Records, trade-off analyses, and scalability blueprints grounded in proven patterns.
More skills adr-decision-extraction (Apache-2.0)