Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add OneWave-AI/claude-skills --skill cowork-sop-writergit clone --depth 1 https://github.com/OneWave-AI/claude-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/onewave-ai/claude-skills/cowork-sop-writer)<a href="https://agentmods.dev/skills/onewave-ai/claude-skills/cowork-sop-writer"><img src="https://agentmods.dev/badge/skills/onewave-ai/claude-skills/cowork-sop-writer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/onewave-ai/claude-skills/cowork-sop-writer"><img src="https://agentmods.dev/badge/skills/onewave-ai/claude-skills/cowork-sop-writer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 17 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00000 | $0.00762 |
| Opus 5 | $0.00000 | $0.00381 |
| Sonnet 5 | $0.00000 | $0.00152 |
| Haiku 4.5 | $0.00000 | $0.00076 |
Grade A, and why
cowork-sop-writer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 35 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cowork SOP Writer
Write procedures the way a good operations manager does: capture how the work is actually done, make the implicit explicit, and write for the new hire on their worst day -- not for the expert who already knows. Input: any record of the process (transcript of someone walking through it, Loom/recording transcript, chat thread, rough notes, or existing scattered docs) and, optionally, an existing SOP folder to match style.
Workflow
- Extract the process. From the source material, pull the sequence of actions, the tools/systems touched, the inputs required to start, the outputs that define done, and who does what. Note every place the narrator said "usually", "unless", "it depends", or "just ask [person]" -- those are the decision points and tribal knowledge the SOP exists to capture.
- Interview the gaps. List what the source did not cover: unnamed systems, missing credentials/access requirements, undefined edge cases, steps that assume knowledge. Ask the user to fill the critical ones; mark the rest
[TO CONFIRM]inline rather than guessing. - Draft. Write the SOP: purpose (one sentence), owner and roles, prerequisites, numbered steps (one action per step, the verb first, the system named, the expected result stated), decision points as explicit if/then branches, exceptions and how to handle them, escalation path, and definition of done. Insert
[SCREENSHOT: what it should show]placeholders where a visual would prevent an error. - Pressure-test. Re-read as the new hire: can every step be executed without asking anyone anything? Any step that requires judgment gets either the judgment criteria written down or an explicit "ask [role]" instruction. Flag steps where the process itself looks fragile (single person dependency, manual copy-paste between systems) in a separate "process risks" note to the owner -- not in the SOP body.
- Library consistency. If an SOP folder was provided, match its template, heading structure, and naming convention. If none exists, propose one (
sop-[team]-[process-name].mdwith a standard header block: owner, last verified date, systems touched) and offer to retrofit existing docs.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 35 lines · 0 tokens per session scan A 96b8540f5ddb
cowork-sop-writer is a skill published in the GitHub repository OneWave-AI/claude-skills (291 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 762 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
give-me-tips
Explains any senpi tip in depth, including Tip: lines in the TUI. Use when the user asks about a tip, what a tipped feature does, or which tips they can see.
suede-onboarding
Suede-affiliated onboarding and activation strategy for first-run sequencing, empty states, setup checklists, activation milestones, time to value, and retention-linked measurement. Use when users sign up but fail to reach first value or the product needs a new first-session flow. NOT FOR: registration optimization…
google-apps-script
Build Google Apps Script automation for Sheets and Workspace. Custom menus, triggers (onEdit / time-driven / form submit), dialogs, sidebars, email batches, PDF export, external API. Use whenever the user wants to automate a Google Sheet, build a Sheets menu / sidebar / dialog, hit a Sheets row from email or a…
google-chat-messages
Send Google Chat messages via incoming webhooks — text, rich cards (cardsV2), threaded replies. TypeScript types, card builder utility, widget reference inline. Use whenever the user wants to post to Google Chat from a script, build a chatbot reply, send a notification card, build a Google Chat webhook integration, or…
parcel-tracking
Track parcels and check delivery status for Australian and international couriers. Searches Gmail for dispatch/shipping emails and provides tracking links for all major Australian couriers including AusPost, StarTrack, Aramex, CouriersPlease, Sendle, Toll, Team Global Express, DHL, FedEx, TNT, Hunter Express, Border…
gws-setup
Set up the Google Workspace CLI (gws) from scratch. Guides through GCP project creation, OAuth credentials, authentication, and installing 90+ agent skills for Claude Code. Use whenever the user wants to set up gws for the first time, configure Google Workspace API access, install the Google Workspace CLI, or…