Content
100%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.
An exemplary lean, single-purpose instruction skill: an executable command, an explicit decision gate, a validation checkpoint on the outbound action with failure handling, and zero padding. All guidance is concrete and every token earns its place.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Every line is load-bearing: the curl command, the DM content rules ('Start the DM with the source channel name and a permalink to the originating thread'), and the reporting constraints ('Say you messaged someone only after reach returns ok: true'). There is no padding and no explanation of concepts Claude already knows, matching the lean-and-efficient anchor; even the one explanatory line ('They haven't seen this conversation') is non-obvious and earns its place. | 5 / 5 |
Actionability | The body provides a copy-paste-ready, complete curl command with environment variables and clearly marked fill-in placeholders, plus concrete rules for DM content and for post-send reporting including the failure path ('If it failed or matched the wrong person, say that instead'). This matches the fully-executable anchor rather than anchor 4, which anticipates minor gaps in the commands or coverage. | 5 / 5 |
Workflow Clarity | Clear three-phase sequence — decide (posture gate: 'Open posture... do the work yourself'; 'Isolated: hand off only when...'), DM (executable command), then report — with an explicit validation checkpoint ('only after reach returns ok: true') and error-recovery guidance for failed or wrong-recipient matches. The single DM is not a destructive or batch operation, and the validation and failure handling present match the explicit-checkpoints-with-feedback anchor; anchor 4 would require missing checkpoints, which is not the case. | 5 / 5 |
Progressive Disclosure | The skill is under 50 lines, needs no external reference files (none exist in the bundle), and is organized into clear sections ('First, can you do it yourself?', 'Ask them by DM', 'Then') that each carry one level of the workflow. The single cross-reference to the sibling request-access skill is clearly signaled and appropriate. Per the scoring note, an under-50-line skill with no need for external references scores 5 with well-organized sections. | 5 / 5 |
Total | 20 / 20 Passed |