Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/686f6c61/alfred-devnpx agentmods add skills/686f6c61/alfred-dev/write-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/skills/686f6c61/alfred-dev/write-adr)<a href="https://agentmods.dev/skills/686f6c61/alfred-dev/write-adr"><img src="https://agentmods.dev/badge/skills/686f6c61/alfred-dev/write-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/686f6c61/alfred-dev/write-adr"><img src="https://agentmods.dev/badge/skills/686f6c61/alfred-dev/write-adr.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00037 | $0.00447 |
| Opus 5.5 | $0.00015 | $0.00179 |
| Sonnet 5.5 | $0.00007 | $0.00089 |
| Haiku 4.5 | $0.00004 | $0.00045 |
Grade A, and why
write-adr scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Escribir un ADR
Resumen
Documenta una decisión arquitectónica significativa en docs/adr/. Un ADR no
es un diario: solo se escribe si hay una elección real (stack, persistencia,
auth, límites, integración).
Proceso
-
Crea el esqueleto con el helper. No inventes el número a mano:
python3 .claude/alfred-continuity.py next-adr "$PWD" --title "Título corto" -
Lee
templates/adr.mdsi existe y el fichero recién creado. -
Rellena, con hechos del repo:
- contexto y restricciones
- al menos dos opciones reales, con pros y contras
- decisión y por qué
- consecuencias y deuda asumida
-
Estado inicial:
propuestohasta que el usuario o la gate de arquitectura lo acepten. Entoncesaceptado. Si más tarde cambia, crea un ADR nuevo que apunte al anterior; no reescribas el viejo. -
Sustituye
<!-- alfred-doc:scaffold -->por<!-- alfred-doc:filled -->. -
Ejecuta
sync-project-docspara refrescar el índice. -
Si la decisión es de diseño, regístrala también con
memory_log_decision. La memoria no sustituye al ADR.
Criterios de éxito
- El fichero está en
docs/adr/ADR-NNN-....md. - Hay al menos dos opciones evaluadas.
- La decisión y sus consecuencias se pueden leer sin el chat.
- El marcador es
filled.
Qué NO hacer
- No abras un ADR por un rename o un ajuste local de
quick. - No dejes
_(pendiente)_en un ADR que declaras cerrado. - No copies un ADR genérico que no menciona el repo.
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.
- 2mo ago First seen · 54 lines · 37 tokens per session scan A 602f37af9365
write-adr is a skill published in the GitHub repository 686f6c61/alfred-dev (115 stars, last pushed 2mo ago), licensed MIT. It adds 37 tokens to every session and 447 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
sdlc-graph-engineering-install
Install graph engineering into a project: take an existing process — skills, a runbook, a README, a CI pipeline, or something the user can only describe out loud — and scaffold it into a working guarded graph with typed nodes, total exit guards, bounded retry loops, a durable run-state file, and a ledger recording…
frontend-design
A guide to building polished web interfaces, including their layout, colors, typography, responsive behavior, and interactive states.
research
A research workflow for investigating a question using multiple sources and producing a cited Markdown report under docs/research/.
architecture
A guide for making high-level software architecture decisions, such as choosing system patterns, databases, and ways to handle growth.
brainstorm
A guided conversation for turning a vague creative idea into a clear proposal. It asks one question at a time and checks what you want, who it is for, how success will be measured, and what limits apply.
oss-strategy
Evaluate an open-source business and community strategy. Triggers on "open source strategy", "OSS 策略".