Content
78%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 a concise, well-structured, self-contained guide with fully executable commands covering install and the common run modes. Its main gaps are the absence of any failure/verification handling around the external command and an underspecified post-run summarization step.
Suggestions
Add one line of fallback guidance for when the context-xray command is not found (e.g., re-run the npx install or verify PATH).
Make the post-run step concrete: point to which part of the generated HTML report contains sessions, context buckets, and warnings so the summary is grounded in actual output.
Trim the defensive clarifications about hosted auth and background servers unless they address a real user confusion.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and assumes competence — no concept explanations, just install and run commands. Two sentences could be trimmed: "It does not need hosted auth or a remote MCP connector to read local transcript files" and "it does not keep a background server running" are defensive clarifications with marginal value. This matches level 4 ('Efficient; minor instances of over-explanation that could be trimmed') rather than level 5, where every token earns its place. | 4 / 5 |
Actionability | Concrete, copy-paste-ready commands are provided for every common case: install ("npx @agent-native/core@latest skills add context-xray --client all"), current thread ("context-xray --open"), picker ("context-xray threads --open"), and trends ("context-xray trends --since 7d --open"). It stops short of level 5 because the post-run instruction ("summarize the report link, sessions analyzed, largest context buckets, warnings, and concrete optimizations") gives no guidance on where to find these details in the report or what to do if the command is missing or fails. | 4 / 5 |
Workflow Clarity | The Setup → Run → summarize sequence is clear and unambiguous with labeled sections and per-command examples. It does not reach level 5 because there are no checkpoints or error-recovery guidance (e.g., verifying the command installed successfully before running), and the final summarize step is a single underspecified instruction. No validation cap applies since the operations are read-only and non-destructive. | 4 / 5 |
Progressive Disclosure | No bundle files exist (references/, scripts/, assets/ are all absent), so the skill is entirely self-contained — and appropriately so for a simple, single-purpose tool skill. The body is a compact ~54-line overview organized into well-labeled sections (Setup, Run) with no content that belongs in separate reference files, matching the simple-skill pattern where well-organized sections alone merit a 5. | 5 / 5 |
Total | 17 / 20 Passed |