Installs into .claude/skills of the current project.
Are you the author of Outcome Drift Audit?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/jgwill-outcome-drift-audit)
---
name: outcome-drift-audit
description: 'Audit Miadi work against explicit STCKin and context-layer outcomes so plausible scaffolds do not pass as successful delivery.'
---
# Outcome Drift Audit
Use this skill when a draft sounds coherent but you need to know whether it actually satisfies the intended outcomes named in the corpus.
## Outcome anchors to check
- STCKin remains part of the STC bot family.
- STCKin keeps its WEST / integration-reflection position.
- Context-layer language preserves Miadi's downstream consumption relationship to AvaLangStack.
- Promotion decisions respect `provenance -> spec -> wiki`.
- Retrieval/context-layer concerns stay distinct from wiki explanation.
- Outputs are durable enough to support later RISE and wiki drafting waves.
## Audit steps
1. Collect the explicit outcome statements from prompts, reports, rispecs, or wiki pages.
2. Gather the current evidence for each outcome.
3. Mark whether the outcome is:
- met
- partial
- asserted-only
- contradicted
4. Explain the drift or missing proof.
5. State the consequence for revision and promotion.
## Output format
Produce a table:
| Outcome | Intended source | Evidence found | Status | Drift or gap | Required action |
| --- | --- | --- | --- | --- | --- |
Then add a short summary:
- strongest drift found
- most important correction
- whether promotion should be blocked
