In this article
engagement-report-conventions
| Field | Value |
|---|---|
| Kind | skill |
| Source | .github/skills/engagement-reporting/engagement-report-conventions |
| Invocation | Loaded on demand by referencing agents |
| Interactive | No |
What it does
Style, terminology, and data-handling rules for engagement reports, their sources, working files, and engagement.yaml. Preloaded by the engagement report generator and its subagents.
When to use it
The engagement report generator and its subagents preload this skill. It holds the rules for report language, names, and data:
- Style: keep status language concise, separate completed work from planned work, and match detail and tone to the configured internal or external audience. The report template owns section structure; this skill owns the language inside it.
- Terminology: when
engagement.yamldefines canonical names for stakeholders, teams, products, or programs, use them consistently. When it defines none, keep the spellings that primary sources use. - Data handling: keep raw content and unneeded personal data out of reports, keep working files apart from final artifacts, and get explicit approval before creating an Outlook draft. These rules do not replace an organization's information-protection, retention, or records-management policies.
Example usage
Instead of writing The team made great progress on deployment, a report
states the evidenced result and status, such as The team completed staging deployment and is validating production prerequisites. If engagement.yaml
maps Project North Star and North-Star to North Star, the report uses
North Star throughout and leaves the source record unchanged.