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.
The body is well-structured with a clear sequenced retrieval workflow, graceful degradation rules, and a concrete output table schema. It is held back from higher scores by the absence of executable search examples and by some repetition of the source/database ordering across sections.
Suggestions
Add 1-2 concrete example queries or API calls for Zotero/Obsidian/IEEE to lift actionability from procedural to copy-paste ready.
Consolidate the source ordering, which is restated in 'Retrieval Order', 'External Search Policy', and 'Key Rules', into one authoritative section to tighten conciseness.
Move the long Tier A/B/C venue lists into a references file (e.g., references/venues.md) to improve progressive disclosure now that the body approaches 300 lines.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Efficient and mostly assumes Claude's competence, with tight bullet structure and minimal over-explanation; minor padded restatements (e.g., external search order repeated across sections) keep it just below a 5. | 4 / 5 |
Actionability | Provides concrete field lists, venue tiers, and an output table schema, but offers no executable commands or example queries (e.g., example IEEE/Zotero search queries) — guidance is procedural rather than copy-paste ready, fitting anchor 3. | 3 / 5 |
Workflow Clarity | Clear sequenced workflow (Step 0a through Step 2) with graceful-degradation checkpoints and explicit skip conditions; lacks explicit validate/retry feedback loops, so it sits at anchor 4 rather than 5. | 4 / 5 |
Progressive Disclosure | Well-organized into clear sections (Purpose, Source Selection, Retrieval Order, Venue Priority, Workflow, Output); no bundle files exist so all content is appropriately inline, with minor opportunities to offload the long venue tier lists to a reference. | 4 / 5 |
Total | 15 / 20 Passed |