Claude Skill MartBrowse skillsWhat is a Skill?
 
← Back to list

protocols.io Integration (Safe Reads & Write Plans)

Reads, validates, and exports protocols.io data against current v3/v4 REST and MCP contracts, while mutations are produced only as non-executing dry-run plans.

Web & APIAdvanced★ 33,030⑂ 3,248AI score 8/10Last updated: Aug 9, 2026

What it does

A safety-first integration for protocols.io, the experimental-protocol sharing platform.

  • Accurate endpoint map: the official docs are still titled "API v3" but mix v3 and v4. This skill tabulates the correct version per operation — list via /api/v3/protocols, get/steps via /api/v4/..., PDF via /view/[id].pdf — and explicitly warns against deprecated patterns.
  • Offline first: credential presence checks, saved-JSON validation, pagination verification, and write planning all run without network or secrets.
  • Explicit --execute gate: network reads only happen when you pass --execute; official HTTPS hosts only, redirects rejected, ambient proxy discovery disabled.
  • Plan-only mutations: create, update, publish/DOI, step deletion, comment deletion, file trash, upload, and org export emit a redacted dry-run plan plus a confirmation phrase — never an actual write.
  • Provenance preserved: title, authors, DOI, version_uri, explicit /vN, license, and fork metadata are kept; archived versions are never silently swapped for /latest.
  • Untrusted remote content: protocol text, comments, filenames, signed upload fields, and error strings are summarized, never obeyed.
  • Rate-limit awareness: 100 req/min per user, PDF 5/min signed in and 3/min anonymous, with bounded idempotent-only retries.

Who it's for

  • Lab researchers publishing or maintaining protocols on protocols.io
  • Grad students who need exact version-pinned protocol citations for reproducibility
  • Research IT staff handling organization exports or the v4 file manager
  • Developers who want guardrails around bearer tokens and irreversible API calls

Examples

  1. Discover public protocols: "find 20 single-cell RNA protocols on protocols.io" — the client first prints the planned URL and bounds, then you re-run with --execute --page-size 10 --max-items 20.
  2. Version-pinned citation: fetch protocol-uri/v2, keep authors/DOI/version_uri intact, and save the PDF only to a fresh 0600 file.
  3. Pre-flight an update: plan_write_request.py --operation update-protocol produces a diff against current state, a permission check, and a note that publishing freezes the version and mints a DOI — then a human decides.
  4. Audit a saved response: run validate_protocol_json.py --require-version to catch broken step GUID ordering or missing attribution metadata.

· · · Install guide · · ·

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/protocolsio-integration folder from the GitHub repo K-Dense-AI/scientific-agent-skills into my ~/.claude/skills/protocolsio-integration/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/K-Dense-AI/scientific-agent-skills.git && mkdir -p ~/.claude/skills && cp -r scientific-agent-skills/skills/protocolsio-integration ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal and go to your working directory.
  2. Clone the repository: git clone https://github.com/K-Dense-AI/scientific-agent-skills.git
  3. Create the skills directory: mkdir -p ~/.claude/skills
  4. Copy just this skill: cp -r scientific-agent-skills/skills/protocolsio-integration ~/.claude/skills/
  5. Confirm Python 3.11+: python3 --version (standard library only, no pip installs needed).
  6. For authenticated reads, sign in to protocols.io, get a token from the Developer resources page, and set PROTOCOLS_IO_ACCESS_TOKEN through your host's secret manager — never paste it into chat or a shell argument.
  7. Verify offline: python3 -B scripts/validate_auth_config.py --require read
  8. Restart Claude Code and ask something like "look up this protocols.io protocol version". Remember that real network calls only occur on commands that include --execute, and write operations stay plan-only by design.
View source on GitHub ↗License: MIT