Content
67%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.
A well-organized router with concrete invocation targets, clear trigger matching, and a valuable conflict-resolution section that deflects non-environment issues. Its main costs are redundancy — the Decision Tree and the long Example Invocations section largely duplicate the Routing Logic — and inconsistent, partly unresolvable reference paths across sibling suites.
Suggestions
Cut the 'Example Invocations' section (or reduce it to a handful of ambiguous cases) — its ~120 lines duplicate the per-skill Triggers and When to Use lists almost one-to-one; the Decision Tree could likewise be merged into the Routing Logic headers.
Normalize every reference to one resolvable format (e.g., a consistent 'suite/skills/<file>.md' path or a local references/ copy), so each Invoke target can actually be followed from this skill.
Fix the section ordering and numbering (16 appears before 15; Decision Tree order diverges from Routing Logic order) and add a one-line tie-break rule for issues matching multiple routes.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The core routing tables are terse trigger lists with no basic-concept over-explanation, but the ~435-line body triplicates the same routing information: the Decision Tree repeats the 16 Routing Logic entries, and ~120 lines of 'Example Invocations' restate the per-skill Triggers as user phrasings. This places it between score 2 ('noticeably verbose; several padded sections') and score 4 ('efficient; minor over-explanation') — the redundancy is substantial but each individual entry is lean, so it sits at the midpoint rather than noticeably below it. | 3 / 5 |
Actionability | Each route ends in a concrete, executable directive: exact file paths ('skills/xcode-debugging.md'), agent names ('Launch spm-conflict-resolver agent'), or slash commands ('/axiom:fix-build', '/axiom:console'). Below score 5 because invocation targets are inconsistently specified — some are bare local paths, others 'See axiom-shipping (skills/testflight-triage.md)' or 'axiom-security/skills/...' pointing into other suites, and none of the referenced files exist in this bundle to verify. | 4 / 5 |
Workflow Clarity | The routing sequence is clear: When to Use → trigger-matched Routing Logic → numbered Decision Tree → explicit 'When NOT to Use' conflict-resolution table with correct-router pointers. Below score 5 because there are no validation checkpoints or tie-breaking guidance for issues matching multiple routes, and the section numbering is disordered (section 16 appears before 15, and Decision Tree order diverges from Routing Logic order), which slightly muddies the prioritization. | 4 / 5 |
Progressive Disclosure | The body is a well-structured router overview with clearly signaled one-level-deep references (an explicit '**Invoke**:' line per route), plus the AXIOM_AUDITOR_INLINE block explaining how to follow agents without Claude Code. Below score 5 because the reference formats are inconsistent across suites (local 'skills/x.md', 'See axiom-shipping (skills/x.md)', '/axiom:console'), no references/ or other bundle files exist in this skill to back the local paths, and out-of-order section numbering weakens navigation. | 4 / 5 |
Total | 15 / 20 Passed |