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 alexclowe/awesome-copilot-cowork-plugins --skill hazard-communication-stylegit clone --depth 1 https://github.com/alexclowe/awesome-copilot-cowork-pluginsWrote 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/alexclowe/awesome-copilot-cowork-plugins/hazard-communication-style)<a href="https://agentmods.dev/skills/alexclowe/awesome-copilot-cowork-plugins/hazard-communication-style"><img src="https://agentmods.dev/badge/skills/alexclowe/awesome-copilot-cowork-plugins/hazard-communication-style/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/alexclowe/awesome-copilot-cowork-plugins/hazard-communication-style"><img src="https://agentmods.dev/badge/skills/alexclowe/awesome-copilot-cowork-plugins/hazard-communication-style.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00039 | $0.00638 |
| Opus 5 | $0.00019 | $0.00319 |
| Sonnet 5 | $0.00008 | $0.00128 |
| Haiku 4.5 | $0.00004 | $0.00064 |
Grade B, and why
hazard-communication-style scanned grade B with 1 finding 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.
Strips warnings and disclaimersmediumAnti-refusal
Omitting safety caveats hides risk from the user and is a common jailbreak preamble.
- Do not moralize about carelessness. Safety writing that blames the reader loses the reader How it starts
The opening of the file, as written. The whole thing — 41 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You write safety material for the people who have to act on it, not for a filing cabinet. When the user is producing anything worker-facing, apply these conventions automatically.
Who you are writing for
A crew member reading this is tired, in a hurry, possibly reading in a second language, and standing up. Safety writing fails at the point of use far more often than it fails on paper.
- Short sentences. One instruction per sentence
- Plain words over technical ones where a plain word exists: "breathing gear" beats "respiratory protection apparatus" on a sign, even if the formal term belongs in the procedure
- Define a technical term the first time it appears, then use it consistently — never rotate synonyms for the same hazard or the same piece of equipment
- Aim for a reading level well below the audience's, not at it
- Avoid idioms and figurative language; they are the first thing to break for a non-native reader
Say what to do, not what to avoid
- Write the required action in the imperative: "Lock out the disconnect and try to start the machine to confirm zero energy." Not: "Machinery should not be serviced while energized"
- Put the action first and the reason second. People stop reading at the first thing that sounds like a policy
- Give the specific trigger — when, where, which task — instead of a general principle. "Before you open the guard" is actionable; "when servicing equipment" is not
- Prohibitions still need a positive alternative: what to do instead, and who to call
Be concrete about consequences without theatrics
- Name the actual mechanism of harm: what the body part contacts, what the energy source is, how quickly it happens. Concrete beats severe
- Do not use fear, statistics-as-shock, or graphic detail as motivation. It reads as manipulation to an experienced crew and it does not change behavior
- Do not moralize about carelessness. Safety writing that blames the reader loses the reader
Structure for the way it will be used
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 · 41 lines · 39 tokens per session scan B 418cb8a6c03a
hazard-communication-style is a skill published in the GitHub repository alexclowe/awesome-copilot-cowork-plugins (17 stars, last pushed 1mo ago), licensed MIT. It adds 39 tokens to every session and 638 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 1 finding (strips warnings and disclaimers). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
custom-daily-brief
Builds a role-tuned morning brief as a Word document covering emails that need replies, today's calendar, Teams mentions and upcoming deadlines, using section presets from an editable config file. Use when the user asks for their daily brief, morning brief, start-of-day summary or custom briefing, or wants to change…
report-attachment-analyzer
Tracks a recurring emailed report end to end. It finds new report emails in the inbox, extracts data from their attachments, appends each reporting period to a trends spreadsheet, and rewrites a summary document with period-on-period deltas and flagged anomalies. Use when the user asks to update or analyse trends for…
commitment-catcher
Scans the last 24 hours of email, Teams chats and meeting transcripts for commitments in both directions, things the user promised to others and things others promised to the user. Maintains a running commitments.md ledger with open, done and overdue statuses, flags overdue items first, and creates To Do entries with…
project-dashboard-builder
Generates a self-contained single-file HTML project dashboard (status banner, milestones, risks heat list, decisions log, links) from the tracker files in /Documents/Cowork/projects/, saved as dashboard- -YYYY-MM-DD.html. Use when the user asks for a project dashboard, a shareable or pinnable status page, an HTML…
inbox-triage
Sweeps the inbox, categorises every message as needs-reply-today, needs-reply-this-week, waiting-on-others, FYI or noise, creates reply drafts in the Drafts folder, and saves a dated triage report to OneDrive. Moves or archives mail only when a rule the user wrote in its rules file says so. Use when the user asks to…
cbo-reviewer
Reviews a proposal, business case, deck or plan in character as a Chief Business Officer archetype, covering commercial model, partnerships, deal structure, market positioning, strategic fit and opportunity cost. Produces a DRAFT review document with a verdict, findings that cite the artifact, risks and five…