engineering-blog-post skill (samber): write posts developers believe

Writes or edits an engineering blog post a skeptical developer believes, with a claim ledger and a credibility check.

Installs: 12.4K; on skills.sh
Stars: 7; on GitHub
Files: 5; besides SKILL.md
Updated: Nov 2026; last repo commit

engineering-blog-post acts as a technical editor: it produces a post an experienced developer reads without their trust alarm going off, and refuses to ship one that would set it off. In draft mode it reads your raw material, picks one of eight post patterns, builds a claim ledger, outlines, drafts and runs a six-check credibility gate. In edit mode it returns a prioritised fix list and rewrites only what you approve.

It triggers on "an engineering blog post", "a debugging write-up", "a public postmortem", "a benchmark post" or editing an existing technical draft. It bundles scripts/scan-draft.sh, a grep-based pass for superlatives, absolute claims and unsourced numbers. SKILL.md says not to use it for tutorials, release notes, migration guides, case studies or talks.

Good for

Engineers turning an incident, rewrite or benchmark into a public post
Editors on a company engineering blog
Product teams whose technical post reads like marketing

Install engineering-blog-post

Claude Code

Copy this prompt.

Install the engineering-blog-post skill in Claude Code. Run this command:

/plugin marketplace add samber/developer-relations-skills
/plugin install developer-relations-skills@developer-relations-skills

Then check that it loaded and tell me how to use it.

2
Paste it into Claude Code. It shows the change and asks before running it.

The README recommends installing the whole collection because its skills reference each other, and engineering-blog-post names other skills in the collection to hand work to. The whole-collection routes add all 93 skills in the repo (counted from the git tree); the repo holds no hooks, agents, commands or MCP server files, and 14 of its skills bundle a script. This skill runs scripts/scan-draft.sh, which needs bash, grep and sed. Pick one route per agent.

From the project README. Source.

What's in the skill

When it loads

Writes or edits a technical blog post a skeptical developer audience will believe - the right post pattern, evidence behind every claim, published trade-offs and limitations, runnable snippets, and no marketing voice. Use whenever the user wants an engineering blog post, a technical article, a "how we built it" or "we rewrote it in X" story, a debugging write-up, a public postmortem, a benchmark post, or a product post that must not read as marketing, and when they ask to review, edit, de-jargon or de-market an existing technical draft - even if they only call it "the article". Not a teaching tutorial - use samber/developer-relations-skills@developer-tutorial. Not release notes, a migration guide, a case study or a talk.

Sections in SKILL.md

Scope check
Two modes
Interview
Workflow
Topic filters
Pattern selection
The claim ledger
Benchmark discipline
Trade-offs and limitations
Code in the post
Register
Blog context
Incident-derived posts
Revise, review, ship
Humanizer pass
Credibility gate

Key facts

Author: Samuel Berthe (named in SKILL.md metadata, LICENSE and the plugin manifest)
License: MIT (LICENSE file, SKILL.md frontmatter, plugin.json and marketplace.json agree)
Skill version: 1.0.5 (SKILL.md metadata)
Installs on skills.sh: 12,425 (Nov 2, 2026)
GitHub stars: 8 for the whole samber/developer-relations-skills repo (Nov 2, 2026)
Last repo commit: Nov 1, 2026
Audits on skills.sh: Gen Agent Trust Hub: pass, Socket: pass, Snyk: pass
What it includes: 6 files in the git tree: SKILL.md, evals/evals.json, 3 files under references/ and scripts/scan-draft.sh.
Needs: bash 3.2+, grep and sed (per the script header)
Part of: The developer-relations-skills plugin: 93 skill folders in the git tree
Works in: The README documents installs for Claude Code, Claude.ai, Codex, Cursor and Gemini CLI, all for the whole set

What it includes

Runs the full workflow in draft mode and returns a fix list in edit mode.
Draft mode runs the full workflow; edit mode returns a fix list and rewrites only approved passages.
Two modes
Asks 11 questions; answers 5 to 7 re-rank the pattern table.
Interview
Requires a post to pass one of five filters, such as counter-intuitive.
A post must pass one of five filters, such as counter-intuitive or surprising capability.
Topic filters
Offers eight post patterns, from Bug Hunt to Benchmarks, ranked by effort and value.
Eight patterns from Bug Hunt to Benchmarks and Test Results, ranked by effort and value.
Pattern selection
Classifies every assertion as measured, shown, sourced or opinion before drafting.
Claim ledger
Classifies the post and lists the hardware, versions, workload and runs to disclose.
Classifies the post and lists what to disclose: hardware, versions, workload, tuning, runs and scripts.
Benchmark discipline
Runs core, clarity, components and consumability passes, then briefs a reviewer.
Core, clarity, components, consumability, then a reviewer brief.
Four revision passes
Requires six checks, such as no unsourced numbers and one honest limitation.
Six checks, such as no unsourced numbers and at least one honest limitation.
Credibility gate
Greps a draft file for superlatives, absolutes, unsourced numbers and undated versions.
Takes a draft file and prints lines matching its superlative, absolute-claim, adoption-claim, certainty, internal-identifier, version and number patterns; options are --quiet and --no-numbers.
Runs code · scripts/scan-draft.sh

From the official docs

Permissions

Low risk

engineering-blog-post is writing instructions plus one bash script that runs grep and sed on a draft file you pass it.

What it can do

Read your data
Create and change things
Delete things
Read-only mode

Watch out for

Incident and benchmark material goes into the conversation
SKILL.md asks about customer names, security detail and internal numbers that must stay unpublished, and has a redaction pass for incident posts. Share only what you can have in your Claude history. github.com
It may save notes to devrel-context.md
SKILL.md tells Claude to keep task and project context in devrel-context.md when memory lives in a file, or in the memory system your tool already uses. Check the file before you commit it. github.com
The whole-set install adds 93 skills, 14 with scripts
The README says to install every skill. The git tree has 93 skill folders, and 14 of them hold .py or .sh files. This skill bundles scripts/scan-draft.sh. The tree holds no hooks, agents or MCP config files. github.com

Technical details

Turning on read-only	SKILL.md describes no read-only mode.
Sign-in and keys	SKILL.md and the bundled script name no account, key or token.
Scopes	Claude writes or edits a post in the conversation scripts/scan-draft.sh runs grep and sed on a draft file you give it Writes devrel-context.md when memory lives in a file

Questions

What does engineering-blog-post do?

It writes or edits a technical blog post for a skeptical developer audience. It picks the right post pattern, makes you back every claim with evidence, requires published trade-offs and checks the draft against a six-point credibility gate.

How do I install it?

Use the install box at the top. The README says to install every skill in the repo, because the skills point to each other. The box shows the whole-set route for each app. There is no Claude app route listed, because the scan script needs a local shell.

What does scan-draft.sh do?

It runs grep over a draft file you give it and prints lines that match its patterns. The script's own comment says every hit is a candidate, not a verdict. It exits 0 for no findings, 1 for findings and 64 for invalid input.

Will it rewrite my whole draft?

Not in edit mode. SKILL.md says an unrequested full rewrite destroys the author's voice, so it returns a prioritised fix list and rewrites only the passages you approve.

Is it safe to install?

The script reads the draft file you pass it. Incident notes, benchmark output and customer details you paste go into your conversation, and SKILL.md has a redaction pass for incident-derived posts. The whole-set install adds 93 skills, 14 of which bundle scripts.

How is it different from developer-tutorial?

This skill writes posts that argue or report. SKILL.md routes a post where the reader learns by following steps to samber/developer-relations-skills@developer-tutorial.

Information

updated Nov 2, 2026

Repository: samber/developer-relations-skills
Folder: skills/engineering-blog-post

samber/developer-relations-skills skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 4

README · github.com
engineering-blog-post SKILL.md · github.com
engineering-blog-post scripts/scan-draft.sh · github.com
LICENSE · github.com