PlugWiki

docs-code-sample-standards Skill (samber): Audit Samples

Makes Claude write a code-sample policy for your developer docs and audit every existing sample into a ranked fix queue.

Installs
12.4K
on skills.sh
Stars
7
on GitHub
Files
9
besides SKILL.md
Updated
Nov 2026
last repo commit

docs-code-sample-standards produces two things: a sample policy and a sample audit. After a scope check and an interview, it inventories the corpus, classifies each sample by execution tier, applies a checklist, ranks fixes, and designs verification and ownership. Claude can run scripts/sample-audit.py to extract and lint every fenced code sample in a docs folder.

SKILL.md says not to use it for writing a quickstart page and names developer-quickstart-guide for that. It bundles 7 reference files, one Python script and an evals file.

Good for

  • Docs engineers whose examples no longer compile
  • SDK teams keeping snippets in step across languages
  • Docs owners checking samples for leaked keys

Install docs-code-sample-standards

Claude Code

  1. 1

    Copy this prompt.

    Install the docs-code-sample-standards 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. 2

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

Install the whole set, not one skill: the README says the skills reference each other, so picking one leaves those pointers going nowhere. The repo tree holds 93 skill folders. Fourteen of them bundle .py or .sh files; docs-code-sample-standards bundles one, scripts/sample-audit.py. The tree holds no hooks, agents or MCP config files.

From the project README. Source.

What's in the skill

When it loads

Defines the policy every code sample in developer documentation must meet and audits an existing sample corpus against it, returning a ranked fix queue. Use whenever the user mentions docs code samples, code snippets in documentation, runnable examples, examples that no longer compile, copy-paste failures, snippet drift after an API change, testing docs examples in CI, SDK snippet parity across languages, sample maintenance ownership, or asks why the examples in their docs do not work - even if they only say the docs are broken. Covers sample anatomy, copy-paste and security safety, execution tiers, single-sourcing from tested code, and language parity. Not for writing a quickstart page - use samber/developer-relations-skills@developer-quickstart-guide.

Sections in SKILL.md

  1. Scope check
  2. Interview
  3. Method
  4. Workflow
  5. Execution tiers
  6. The six policy decisions
  7. Invocation examples
  8. Audit output
  9. Sample audit - <corpus>, <date>
  10. P0 - ship-blocking (11)
  11. Fix queue
  12. Pass threshold
  13. KPIs
  14. Failure modes
  15. References

Key facts

Author
Samuel Berthe (named in SKILL.md metadata and plugin.json)
License
MIT (SKILL.md, plugin.json and the LICENSE file agree)
Version
1.0.2 (SKILL.md metadata)
Installs on skills.sh
12,362 (Nov 2, 2026)
GitHub stars
7 for the whole samber/developer-relations-skills repo (Nov 2, 2026)
Last repo commit
Nov 1, 2026
What it includes
10 files in the git tree including SKILL.md. 1 Python script.
Needs
SKILL.md lists nothing to install. The bundled scripts/sample-audit.py needs python3 and imports only the standard library
Part of
The developer-relations-skills plugin: 93 skill folders in the git tree
Works in
Claude Code, Claude.ai, Codex, Cursor and Gemini CLI, as the README documents

What it includes

  • Routes elsewhere when the task is not about the sample corpus

    Read · Scope check

  • Asked one at a time until surfaces, languages and tier-1 scenarios are known

    Read · Interview

  • Three named methods, cited in the deliverable

    Read · Method

  • Inventory, classify by tier, check, rank, design verification, assign ownership

    It points to references for verification patterns and governance.

    Write · Workflow

  • Every sample gets exactly one tier, and tiers the constraints rule out are deleted

    Write · Execution tiers

  • Decides each explicitly and writes them into the policy

    Write · Six policy decisions

  • A short report with a P0 list and a fix queue

    P0 covers items such as credential leaks.

    Write · Sample audit report

  • Walks a docs folder and lints every fenced code block

    Formats are tsv, json or summary; flags include credential-shaped literals and unguarded destructive commands.

    Runs code · sample-audit.py

From the official docs

Permissions

Low risk

docs-code-sample-standards is planning instructions plus one script, and it may save notes to devrel-context.md. Our advice: check that file before you commit it.

What it can do

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

Watch out for

  • 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

  • It can run a Python script on your docs

    sample-audit.py opens the files in the docs folder you give it and prints one row per fenced code block. It flags patterns such as key-shaped strings and unguarded destructive commands but does not judge whether a sample is correct. 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. The tree holds no hooks, agents or MCP config files. github.com

Technical details
Turning on read-onlySKILL.md describes no read-only mode.
Sign-in and keysSKILL.md names no account, key or token.
Scopes

  • Claude writes the plan or report in the conversation

  • May save notes to devrel-context.md when memory lives in a file

  • Can run python3 scripts/sample-audit.py against a docs folder you name

Questions

What does docs-code-sample-standards do?

It makes Claude define the policy every code sample in your docs must meet, then audit your existing samples against it. The audit ends in a ranked fix queue.

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, so the box shows whole-set routes.

Does it run my code samples?

Not by itself. The bundled script lints fenced blocks in Markdown files and reports pattern defects. Its header says judgement checks, such as whether a symbol still exists, are out of scope and need the product.

Can it find leaked keys in docs?

It can flag credential-shaped literals. The header says each flag is a candidate to confirm, because a page about key rotation can legitimately contain a key-shaped string.

Can it write my quickstart?

No. SKILL.md names developer-quickstart-guide for that.

Is it safe to install?

Claude can run one Python script that reads Markdown files under a docs folder you name and prints a report. Claude may also save notes to a devrel-context.md file. See the permissions section.

Information

updated Nov 2, 2026

Repository
samber/developer-relations-skills
Folder
skills/docs-code-sample-standards

samber/developer-relations-skillsskills.sh

Sources

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

Show all 7
  1. README · github.com
  2. marketplace.json · github.com
  3. plugin.json · github.com
  4. docs-code-sample-standards SKILL.md · github.com
  5. LICENSE · github.com
  6. skills.sh listing · skills.sh
  7. docs-code-sample-standards scripts/sample-audit.py · github.com

Updated Nov 2, 2026.