lean-changelog

スター4

フォーク0

更新日2026年7月18日 15:25

Turn findings-diff results into release-notes prose — short, honest "sustainability improvements in this release" sections for changelogs, release announcements, and internal newsletters. Use this skill whenever the user wants to communicate efficiency progress in a release, asks for changelog/release-notes text about green improvements, or wants findings-diff output made publicly legible. Part of Lean Agentic AI Skills; a transformer from diff results to publishable prose.

インストール

Codex または Claude でインストールこの Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。

ソース

navveenb

navveenb/lean-agentic-ai-skills

GitHub リポジトリを開く Creator のリポジトリを見る

ダウンロード

Green Changelog

Transformer skill — the public-legibility layer. Input: findings-diff output (or two findings files to diff first), plus the publication venue (public changelog, internal notes, blog). Output: a short prose section ready to paste.

Progress that isn't communicated doesn't compound — teams stop investing in what leadership and users never see. But sustainability claims in public are exactly where inflated language does the most damage, so this skill writes with the module's tightest wording discipline.

Writing rules

Lead with the countable. "This release resolved 3 efficiency findings, including converting all article images to AVIF (image transfer per page roughly halved — sizes in the release assets)." Counts, formats, and byte sizes from evidence are quotable; energy and carbon are not.
Name the change, not the virtue. "Search results are now cached for repeat queries" beats "we care deeply about the planet". The venue's readers can judge significance from specifics; adjectives invite skepticism.
The unmeasured stays visible in public too — a light-touch version: "…we haven't instrumented energy measurement yet; transfer sizes are our proxy." One clause, not a confession paragraph — but present. Public receipts with humility age well; ones without age badly.
Regressions are mentioned when material. If the diff shows a new high-severity finding shipped in this release, an honest changelog says what's tracked for next release. Silence on regressions converts the changelog from record to marketing.
Tone fits the venue — public changelog: 2-5 bullets, plain; internal newsletter: can carry the posture table and shout-outs; blog post: the story of one fix in depth beats a list of ten.
Never retro-claim: only changes present in the diff between the two audited states are claimable in that release's section.

Output shape

A ready-to-paste markdown section titled for the venue's convention (e.g. "Efficiency & sustainability"), plus a one-line summary suitable for a release-tweet, both derived strictly from the diff data handed in.

Cost signal rendering

Countable cost claims are the only quotable numbers. Where the release resolved findings whose cost_signal.observed was measured before/after, those deltas ARE quotable in the changelog ("prompt tokens per call: 370 → 110"). Energy/carbon claims stay qualitative.

Not this skill's job

Computing the diff (findings-diff), full reports (lean-report-writer), ESG/compliance text (explicitly out — route to lean-report-writer's clearly-labeled ESG-flavored mode, and even that is not assurance-grade).

このリポジトリの他の Skills

同じリポジトリ

agentic-loop-auditor

navveenb/lean-agentic-ai-skills

Audit agent frameworks and agentic workflows for token-burning loop pathologies — unbounded retries, excessive planning depth, tool-call thrashing, context snowballing, missing loop budgets and kill switches. Use this skill whenever the user shares agent configs or traces (LangChain/LangGraph, CrewAI, AutoGen, custom loops), complains an agent is slow or expensive, or is designing any autonomous/multi-step agent. Part of Lean Agentic AI Skills; emits lean-findings.json.

2026-07-184

ai-eval-efficiency-audit

navveenb/lean-agentic-ai-skills

Audit AI evaluation infrastructure for wasted compute — full benchmark suites re-run on unchanged cases, LLM-as-judge grading without caching, oversized judge models, redundant eval passes per commit, and missing result reuse. Use this skill whenever the user shares eval harness configs or CI eval steps, complains that evals are slow or expensive, mentions LLM-as-judge costs, or runs benchmark suites on every change. Part of Lean Agentic AI Skills; emits lean-findings.json.

2026-07-184

api-efficiency-audit

navveenb/lean-agentic-ai-skills

Audit API design for chattiness and over-fetching — REST endpoints requiring N calls per screen, GraphQL resolvers with N+1 fan-out, default-everything payloads, missing pagination and field selection, provider-side polling where webhooks serve, and uncompressed responses. Use this skill whenever the user shares API schemas (OpenAPI, GraphQL SDL), resolver code, endpoint designs, or asks why clients make too many calls or payloads are heavy. Part of Lean Agentic AI Skills; emits lean-findings.json.

2026-07-184

caching-strategy-designer

navveenb/lean-agentic-ai-skills

Design caching layers so work is done once, not per request — HTTP/CDN caching, application-level caches (Redis/Memcached), computed-result caching, and cache-invalidation strategy. Use this skill whenever the user wants to add caching, asks why the same work is recomputed, designs a read-heavy system, or an audit skill (db-efficiency-audit, web-carbon-audit) flags uncached repeated work. Part of Lean Agentic AI Skills; emits a cache design plus lean-findings.json.

2026-07-184

carbon-aware-scheduler

navveenb/lean-agentic-ai-skills

Design time-shifting of flexible workloads (batch jobs, CI, training, backups, crons) to hours and regions when grid electricity is cleanest, using carbon-intensity signals like Electricity Maps, WattTime, or the Carbon Aware SDK. Use this skill whenever the user mentions carbon-aware computing, scheduling flexible/deferrable jobs, demand shifting, running workloads when energy is clean, or asks how to reduce the carbon intensity (not just amount) of their compute. Part of Lean Agentic AI Skills; emits schedule designs plus lean-findings.json.

2026-07-184

carbon-budget-designer

navveenb/lean-agentic-ai-skills

Design per-feature and per-service efficiency budgets — page-weight budgets, token budgets per AI feature, query budgets per endpoint, GPU-hour budgets per training project — with owners, review checkpoints, and enforcement hooks. Use this skill whenever the user wants sustainability targets for teams, asks how to keep efficiency from regressing organizationally, mentions performance-budget-style governance for carbon/energy, or is setting engineering standards. Part of Lean Agentic AI Skills; an advisor emitting budget definitions plus lean-findings.json.

2026-07-184

name

lean-changelog

description

Green Changelog

Writing rules

Lead with the countable. "This release resolved 3 efficiency findings, including converting all article images to AVIF (image transfer per page roughly halved — sizes in the release assets)." Counts, formats, and byte sizes from evidence are quotable; energy and carbon are not.
Name the change, not the virtue. "Search results are now cached for repeat queries" beats "we care deeply about the planet". The venue's readers can judge significance from specifics; adjectives invite skepticism.
The unmeasured stays visible in public too — a light-touch version: "…we haven't instrumented energy measurement yet; transfer sizes are our proxy." One clause, not a confession paragraph — but present. Public receipts with humility age well; ones without age badly.
Regressions are mentioned when material. If the diff shows a new high-severity finding shipped in this release, an honest changelog says what's tracked for next release. Silence on regressions converts the changelog from record to marketing.
Tone fits the venue — public changelog: 2-5 bullets, plain; internal newsletter: can carry the posture table and shout-outs; blog post: the story of one fix in depth beats a list of ten.
Never retro-claim: only changes present in the diff between the two audited states are claimable in that release's section.