In this article
engagement-report-council-critique
| Field | Value |
|---|---|
| Kind | skill |
| Source | .github/skills/engagement-reporting/engagement-report-council-critique |
| Invocation | Invoked directly as /engagement-report-council-critique; model invocation is disabled, so agents do not load it |
| Interactive | No |
What it does
Run one independent Council critique against engagement-report research evidence. Use as the manual separate-session fallback when independent critic agents are unavailable.
When to use it
Use this skill to run an independent Council critique manually when isolated subagent runs are unavailable. Run it in at least two separate model sessions with the same draft and research evidence, but do not expose one critique to another.
Use the Engagement Report Reviewer when only one independent review is needed.
How to use it
- Prepare the draft, research findings, coverage summary, audience, and selected template.
- Start a separate model session for each critic.
- Run
/engagement-report-council-critiquewith the prepared inputs, the reporting date, the report-type slug, and a unique critic-run slug. The skill derives the confined critique path. Confirm effective-ignore protection for.working/before the invocation. - Save each critique independently.
- Provide at least two completed critiques to the Council Arbiter.
Example usage
/engagement-report-council-critique report-date=2026-08-28 report-type=weekly critic-run=critic-2
audience="customer steering committee" draft=[draft content]
research=[normalized findings] coverage=[coverage record] template=[selected template]
Effective-ignore protection for .working/: confirmed
The skill returns evidence-linked accuracy, completeness, proportion, directionality, privacy, terminology, and continuity findings without rewriting the report.