Content
85%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured discovery/audit skill with strong workflow clarity and actionability. The adaptive workflow branching (broad audit vs. cleanup planning vs. targeted investigation) is particularly effective, and the tool usage guidance is specific and concrete. The main weakness is minor verbosity—the opening paragraph and some of the Important Context section could be tightened without losing clarity.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is mostly efficient but includes some unnecessary framing ('You're using a skill that will guide you through...') and explanatory text that Claude doesn't need (e.g., explaining what 'launched' means in the Important Context section is borderline—useful domain knowledge but somewhat verbose). The workflow steps are well-structured but could be tightened in places. | 2 / 3 |
Actionability | The skill provides concrete, specific guidance on which MCP tools to use, with exact parameter names (projectKey, inactiveDays, includeOnly), specific filter values (state types, staleness categories like never_requested/inactive_30d/launched_no_changes), and clear decision trees for different user goals. While there's no executable code, this is an instruction-only skill where the actionability comes from precise tool usage guidance. | 3 / 3 |
Workflow Clarity | The 6-step workflow is clearly sequenced with adaptive branching based on user goals. It includes validation checkpoints (Step 5's removal readiness with safe/caution/blocked verdicts), explicit decision points (Step 4's categorization), and clear handoffs to other skills for action. The workflow handles the discovery-only scope well with appropriate guardrails. | 3 / 3 |
Progressive Disclosure | The skill provides a clear overview with well-signaled one-level-deep references to flag-health-signals.md and removal-readiness-checklist.md. It also cross-references sibling skills (flag-cleanup, flag-targeting) appropriately. Content is well-split between the main workflow and detailed reference materials, though bundle files weren't provided to verify the references exist. | 3 / 3 |
Total | 11 / 12 Passed |