Traces requirements to stories bidirectionally and reports coverage gaps and drift.
npx skills add https://github.com/DATA-AI-XYZ/Tandem --skill plan-drift-data-ai-xyz
Please help me install this Agent Skill. Skill: plan-drift Source: https://github.com/DATA-AI-XYZ/Tandem/tree/main/skills/plan-drift Command: npx skills add https://github.com/DATA-AI-XYZ/Tandem --skill plan-drift-data-ai-xyz
High-intent search queries and answers about installing and using this skill.
Run a plan-drift pass over the whole repo or a named epic. It inventories every PRD key requirement, epic In-scope bullet, and feature acceptance criterion, then traces each through the existing linkage chain to find stories that cover it, reporting any gaps.
The fidelity check compares each story's acceptance criteria against the requirement its PRD reference resolves to. It flags scope-adds, scope-drops, reinterpretations, and stale source requirements as advisory findings for human review.
No, it is strictly read-only. The only file it writes is a dated report under 41-Reports/audits/, and it never edits stories, PRDs, epics, features, or any status field.
Yes, you can scope the run to a named epic, which restricts the requirement inventory to that epic's PRD, its In-scope bullets, and the features and stories under it. The whole-repo run is the thorough default.
The report explicitly states in its Scope section that the resolved requirement folder was empty rather than silently reporting a clean pass. An empty inventory is treated as a failure mode because it looks identical to full coverage.
petrSimonidesXartValidate Discovery artefacts against governance rules and required frontmatter fields.
gaiastudio-aiAnalyze acceptance criteria and test suites to document coverage gaps.
heymishyReview story artefacts and emit severity-tagged findings with diffs.
metasaverDefine a standard template for epics and user stories extracted from PRDs.
