Docs Pipeline Dry Run

VerifiedCaution

Runs the diff and planning phases of the weekly docs pipeline in read-only mode. Reports what a full run would do without writing to src/ or advancing state.

Sby Skills Guide Bot
DocumentationAdvanced
7/23/2026
Claude Code
#docs-pipeline#dry-run#diff-analysis#read-only#planning

Recommended for

Automated summary

Runs phases 0 and 1 of the weekly docs pipeline in read-only mode: clones the product repos, diffs them since the last processed commit, plans documentation work items, and produces a report without modifying anything.

Strengths

  • Enforces side-effect safety: no writes under src/, no commits, state.json never advanced
  • Emits inspectable JSON artifacts (diff.json, plan.json) in a dated, gitignored run directory
  • Includes read-only validators (sidebar, link checks) with counts summarized by type
  • Gives a full preview of what a /docs-weekly run would actually do

Limitations

  • Relies on repo-specific orchestration files (docs-diff.js workflow, repos.json, state.json)
  • Analytics signals are not yet available (M4), which limits prioritization quality
  • Covers only phases 0 and 1: no writing, screenshots, or full validation
When to use it

Use it to inspect and preview the impact of the next weekly docs run before triggering the full pipeline.

When not to use it

Do not use it to actually write or publish documentation, since the skill is strictly read-only and does not advance state.

Security analysis

Caution
Quality score85/100

The skill uses network, filesystem, agent spawning, and local script execution, which are powerful capabilities, but its instructions are explicitly read-only toward source and state files and prohibit commits/pushes. No destructive, exfiltrating, or obfuscated behavior is present.

Findings
  • •Clones three product repositories over the network into a temp workspace.
  • •Executes local Node scripts (`validate-sidebar.js`, `check-links.js`) whose contents are not included.
  • •Spawns an agent and passes it the full diff contents.
  • •Deletes `RUN_DIR/sources/` during cleanup.
  • •No allowed-tools are declared, so effective tool permissions are unclear.
How this analysis is made

Examples

Preview weekly docs impact
Do a dry run of the docs pipeline for today: diff the three product repos since the last processed commit and show me the planned work items.
Check docs health before a full run
Run docs-dry-run and report per-repo user-facing changes plus the current sidebar and link validator counts.
Inspect planned screenshots
Dry-run the docs pipeline and list the screenshot requests the plan generated, along with the deferred items.

name: docs-dry-run description: Phases 0+1 of the docs pipeline, read-only. Clones the three product repos into a temp workspace, diffs them since the last processed commit, plans the documentation work items, and reports. Writes nothing under src/ and never advances state.json.

Docs pipeline dry run (Phases 0+1, read-only)

Run the diff and planning phases of the weekly docs pipeline and report what a full run WOULD do. Hard rules for this skill:

  • Never edit anything under src/, automation/state.json, or automation/*-baseline.json.
  • Never commit, push, branch, or open a PR.
  • Write only under automation/runs/<date>/ (gitignored).

Steps

  1. Prepare. Let DATE = today as YYYY-MM-DD, RUN_DIR = automation/runs/<DATE>. Create RUN_DIR. Read automation/state.json and automation/repos.json.

  2. Phase 0, diff. Invoke the Workflow tool with scriptPath: ".claude/workflows/docs-diff.js" (by path, not by name, see DESIGN.md 5.2.1) and:

    args: {
      date: <DATE>,
      runDir: "automation/runs/<DATE>",
      repos: [
        { key, product, url, lastSha: state.lastSha[key], docsTargets, productMapPath: "automation/product-map.json" only for teamspro }
        ... one entry per repo in repos.json
      ]
    }
    

    Save the returned object to RUN_DIR/diff.json.

  3. Phase 1, plan. Spawn one doc-gap-analyst agent. Give it: the full contents of RUN_DIR/diff.json, the writer cap (12), and note that analytics signals are not yet available (M4). Ask for the structured work plan (workItems, deferred, attention, stats). Save it to RUN_DIR/plan.json.

  4. Validators (read-only). Run node automation/scripts/validate-sidebar.js --json and node automation/scripts/check-links.js --json and capture their summaries (counts by type, do not paste hundreds of findings).

  5. Report. Print a readable summary for the user:

    • Per repo: commits analyzed (diff range), user-facing changes found, anything in attention (unmapped modules, suite-wide changes), fallback windows or clone errors.
    • The ranked work items: id, action, product, pages, priority, one-line rationale. Then the deferred list.
    • Screenshot requests the plan generated (they would be captured in Phase 3).
    • Current docs health from the validators (counts only).
    • Close with: what a full /docs-weekly run would now do (write, screenshot, validate, open PR) and that state.json was NOT advanced.
  6. Cleanup. Delete RUN_DIR/sources/ (the temp clones). Keep the JSON artifacts for inspection.

From the file to a first result

The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.

View the source and supporting files

This download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.

1. Download the file

Download SKILL.md

From the root of a test project, save the file as .skill-downloads/docs-pipeline-dry-run-mrx3xwgg/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.

Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.

mkdir -p '.skill-downloads/docs-pipeline-dry-run-mrx3xwgg' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/docs-pipeline-dry-run-mrx3xwgg/download' --output '.skill-downloads/docs-pipeline-dry-run-mrx3xwgg/SKILL.md.part' && mv '.skill-downloads/docs-pipeline-dry-run-mrx3xwgg/SKILL.md.part' '.skill-downloads/docs-pipeline-dry-run-mrx3xwgg/SKILL.md'

2. Check prerequisites

Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.

No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.

The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.

3. Load the skill

After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.

Claude Code
test -f '.skill-downloads/docs-pipeline-dry-run-mrx3xwgg/SKILL.md' && test -s '.skill-downloads/docs-pipeline-dry-run-mrx3xwgg/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/docs-dry-run' && cp -R '.skill-downloads/docs-pipeline-dry-run-mrx3xwgg' '.claude/skills/docs-dry-run'

OpenSkills

OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.

Official OpenSkills installation
openskills install './.skill-downloads/docs-pipeline-dry-run-mrx3xwgg'

openskills read 'docs-pipeline-dry-run-mrx3xwgg'

Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.

For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.

openskills sync

Manual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.

4. Try a first task

Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:

Read .skill-downloads/docs-pipeline-dry-run-mrx3xwgg/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.

Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.

Related skills