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.
npx skills add kaisa-kucherenko/claude-code-flow --skill adrgit clone --depth 1 https://github.com/kaisa-kucherenko/claude-code-flowWrote 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/skills/kaisa-kucherenko/claude-code-flow/adr)<a href="https://agentmods.dev/skills/kaisa-kucherenko/claude-code-flow/adr"><img src="https://agentmods.dev/badge/skills/kaisa-kucherenko/claude-code-flow/adr/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/skills/kaisa-kucherenko/claude-code-flow/adr"><img src="https://agentmods.dev/badge/skills/kaisa-kucherenko/claude-code-flow/adr.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 | $0.00142 | $0.04369 |
| Opus 5.5 | $0.00057 | $0.01748 |
| Sonnet 5.5 | $0.00028 | $0.00874 |
| Haiku 4.5 | $0.00014 | $0.00437 |
Grade A, and why
adr 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.
How it starts
The opening of the file, as written. The whole thing — 315 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Architecture Decision Record (ADR)
"Beads on a necklace" — sequential isolated decisions that build on each other. Each decision narrows the space for the next one.
What this skill produces
Default — single file {feature}-decisions.md in the project root. This file is the source /spec cuts phases from: it records the decisions and their reasons, not the order of work. Works well for features up to ~15 decisions and ~800 lines.
The document opens with two things, before any decision: a short "How it works" (the whole mechanism in 5-10 lines, as it stands after the decisions made so far — rewritten as decisions land) and a status table (number, title, DECIDED / PENDING / DROPPED). A reader who lost the thread re-anchors here, not by scrolling the decisions.
Companion visual — the storyboard "Life of one X". The main visual of the ADR, built together with the decision tree in Phase 3 and kept alive through Phase 4 (see there). The tree is a list; the storyboard is the story the list is about — it is what makes the whole picture hold at the end instead of falling apart into sixteen items.
For larger features — split into per-decision directory docs/adr/{domain}/:
- One README.md (index + status table + reading order)
- One file per decision or per tightly-grouped decision cluster (e.g.
01-platform-install.md,02-schema-source-of-truth.md, ...,11-documentation-rollout.md) - Logical grouping, not strict 1-to-1 (combine tightly related decisions into one file when they share rationale, e.g.
3 + 4→ single file; keep large independent decisions like7standalone) - Cross-references via markdown links (
[02-schema-source-of-truth.md](02-schema-source-of-truth.md)) instead of inlineper 7text
When to split — one scale: ~800 lines / ~25 KB or 15+ decisions. Past ~800 lines the
Read tool hits its 25k-token limit on Cyrillic content and humans stop reviewing coherently;
past 15 decisions the file keeps growing, so split early. Two further triggers regardless of
size: unrelated domains in one ADR (DB + secrets + CI — targeted edits and smaller review
passes), and a project that already uses docs/adr/{domain}/. Below the scale, stay in one
file — a split adds navigation for nothing, and a prototype whose decisions will consolidate
or die is not worth the carpentry.
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 Changed 48295d2fede8
- 1mo ago First seen · 315 lines · 142 tokens per session scan A 5ce197b0e61c
adr is a skill published in the GitHub repository kaisa-kucherenko/claude-code-flow (24 stars, last pushed 2d ago), licensed MIT. It adds 142 tokens to every session and 4,369 once invoked, about $0.0006 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 2026-09-19.
adr compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| adrkaisa-kucherenko/claude-code-flow | Skill | 24 | 2d ago | MIT | A | 142 | Claude Code |
| imagegenopenai/codex | Skill | 133k | yesterday | Apache-2.0 | A | 113 | Codex |
| playwright-climicrosoft/playwright | Skill | 99k | yesterday | Apache-2.0 | A | 19 | Claude Code |
| local-ai-agentsmicrosoft/ai-agents-for-beginners | Skill | 79k | 4d ago | MIT | A | 200 | Claude Code, Codex |
| systematic-debuggingobra/superpowers | Skill | 313k | 3d ago | MIT | A | 21 | Claude Code |
| next-cache-components-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 95 | Claude Code |
| next-partial-prefetching-adoptionvercel/next.js | Skill | 143k | yesterday | MIT | A | 103 | Claude Code |