Content
82%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is highly actionable with concrete commands and clear branching, and is well structured; its main gaps are repeated behavioral admonitions and a script reference whose target is absent from the bundle.
Suggestions
Consolidate the repeated honesty/no-false-success admonitions into one statement in 'Antwortstil und Grenzen' to remove redundancy.
Add an explicit validate→fix→retry loop for the start failure case rather than only 'explain the cause'.
Ensure the referenced scripts/ccv-projects.mjs ships in the bundle (or document where it lives) so the one-level reference resolves.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly lean and actionable, but behavioral constraints are repeated (no false success / don't invent an address stated in both Fähigkeit zwei and Antwortstil), which could be trimmed. | 4 / 5 |
Actionability | Copy-paste-ready commands ('node scripts/ccv-projects.mjs <list|probe|start> [dir]') with exact output handling, a sample address line, and explicit success/failure branches covering the common cases. | 5 / 5 |
Workflow Clarity | Clear list→select→start→respond sequence with readiness and already-running checks delegated to the script plus a failure branch, but there is no explicit validate→fix→retry feedback loop and error recovery stops at 'explain the cause'. | 4 / 5 |
Progressive Disclosure | Well-organized into labeled sections (Begleitskript, Fähigkeit eins/zwei/drei, Antwortstil) with a single one-level-deep script reference and no nested refs, but the referenced scripts/ccv-projects.mjs is not present in the bundle. | 4 / 5 |
Total | 17 / 20 Passed |