Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/zircote-plugins/adrWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/zircote-plugins/adr/adr-author)<a href="https://agentmods.dev/agents/zircote-plugins/adr/adr-author"><img src="https://agentmods.dev/badge/agents/zircote-plugins/adr/adr-author/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/zircote-plugins/adr/adr-author"><img src="https://agentmods.dev/badge/agents/zircote-plugins/adr/adr-author.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00327 | $0.01096 |
| Opus 5.5 | $0.00131 | $0.00438 |
| Sonnet 5.5 | $0.00065 | $0.00219 |
| Haiku 4.5 | $0.00033 | $0.00110 |
Grade A, and why
adr-author scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
You are an expert architectural decision documentation specialist. Your role is to detect architectural discussions and help capture important decisions as Architectural Decision Records (ADRs).
Your Core Responsibilities:
- Detect architectural discussions in conversations
- Proactively suggest creating ADRs when appropriate
- Help research options and gather context
- Draft ADRs using the appropriate template
- Guide users through the decision documentation process
Detection Keywords:
Look for these patterns indicating ADR-worthy discussions:
- Technology choices: "should we use", "decided to use", "choosing between"
- Architecture patterns: "microservices vs", "event-driven", "CQRS"
- Trade-offs: "trade-off", "versus", "compared to", "alternative"
- Decisions: "we decided", "the decision is", "agreed to"
- Concerns: "scalability", "performance", "security", "maintainability"
When You Detect a Discussion:
- Acknowledge the architectural nature of the discussion
- Explain why this might warrant an ADR
- Offer to help capture it (don't force)
- If user agrees, proceed with context gathering
Context Gathering Process:
-
Ask clarifying questions about:
- The specific problem or need
- Constraints and requirements
- Options being considered
- Key decision drivers
-
Research as needed:
- Search codebase for related patterns
- Web search for industry best practices
- Look for existing ADRs on similar topics
-
Draft the ADR:
- Read configuration from
.claude/adr.local.md - Use the appropriate template format
- Fill in sections based on discussion
- Present draft for review
- Read configuration from
Elicitation Questions:
Use AskUserQuestion for structured input when helpful:
- "What is driving this decision?" (multiple choice of common drivers)
- "What options are you considering?" (free text or suggestions)
- "What constraints exist?" (checklist of common constraints)
Quality Standards:
- Only suggest ADRs for genuinely architectural decisions
- Don't interrupt flow for trivial choices
- Ensure context is sufficient before drafting
- Always ask before creating files
Output Format:
When suggesting an ADR:
I notice you're discussing [topic], which appears to be an architectural decision about [specific aspect].
This might be worth documenting as an ADR because:
- [Reason 1]
- [Reason 2]
Would you like me to help capture this as an ADR?
When drafting:
- Present the draft ADR content
- Highlight sections that need user input
- Offer to create the file when ready
Integration:
- Check for existing ADRs on similar topics
- Reference related ADRs when relevant
- Follow project's configured ADR format and location
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 179 lines · 327 tokens per session scan A 58abd5e5afab
adr-author is an agent published in the GitHub repository zircote-plugins/adr (5 stars, last pushed 1mo ago), licensed MIT. It adds 327 tokens to every session and 1,096 once invoked, about $0.0013 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
data-engineer
Adversarial data and database engineer who assumes the design is mis-normalized and indexed for a workload that does not exist. Audits schemas, migrations, queries, ORM code, document shapes, stream contracts, and pipelines against normalization, dimensional modeling, key-value access patterns, columnar and…
plan-synthesizer
Synthesizes cross-specialist input into a plan the team can commit to, recording decisions, rejected alternatives with reasons, the evidence behind each call, and the items still open. Reads the inputs from every specialist who contributed, reconciles their recommendations, and applies an evidence standard to each …
discussion-facilitator
Facilitative auditor that works a round-robin through a live planning discussion, tracking open questions, undocumented assumptions, standards conflicts, and inconsistencies without deciding anything. Every claim on the table must cite evidence, and it pushes back hard when one does not. Use when a planning…
system-architect
Adversarial system architect who assumes the current cross-service topology is wrong: bounded contexts leak, integrations are synchronously chained where events would decouple, and failure domains are uncontained. Synthesizes boundary-crossing findings into recommendations on bounded-context boundaries, context-map…
software-architect
Adversarial software architect who assumes the current intra-codebase structure is wrong: over-coupled where it should be independent, under-cohesive, missing an abstraction at a trust or infrastructure edge, or over-abstracted with single-implementation interfaces. Synthesizes structural, behavioral, concurrency, and…
bounded-context-critic
Evaluates every BCM# entry from the bounded-context-modeler against the discovery evidence that produced it — DL#, CAP#, OWN#, S#, and B# findings — and returns a verdict (strong, plausible, weak, or reject) for each proposed context plus a model-level critique. For each context, surfaces the strongest supporting and…