Content
57%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill provides genuinely useful, hard-won knowledge about Todoist API quirks (priority inversion, ID format issues, filter syntax limitations) that Claude wouldn't know otherwise. However, it suffers from significant internal repetition that inflates token cost, lacks validation/verification steps for destructive and batch operations, and could benefit from consolidating repeated content and adding concrete invocation examples.
Suggestions
Consolidate repeated information (priority inversion, due date handling, ID formats) into single authoritative sections referenced from workflows, rather than restating in multiple places.
Add explicit validation/verification steps after destructive operations (DELETE_TASK, DELETE_SECTION) and batch operations (BULK_CREATE_TASKS) — e.g., 'After bulk create, call GET_ALL_TASKS with project_id filter to verify all tasks were created.'
Add at least one concrete, copy-paste ready MCP tool invocation example showing exact parameter structure (especially for BULK_CREATE_TASKS where the due field structure differs from CREATE_TASK).
Consider splitting the detailed pitfalls and parameter reference into a separate bundle file to keep SKILL.md as a concise overview with workflow sequences.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill contains substantial useful detail (pitfalls, parameter docs, ID format quirks) that Claude wouldn't inherently know, but there's significant repetition — priority inversion is explained three times (in Create Tasks, Known Pitfalls, and implicitly in filter syntax), due date handling is repeated in both the Create Task section and Common Patterns, and ID format issues appear in multiple places. The quick reference table largely duplicates information already covered in the workflows. | 3 / 5 |
Actionability | The skill provides concrete tool names, specific parameter names with formats, and clear tool sequences for each workflow. However, there are no executable code examples or copy-paste ready MCP call examples showing exact JSON payloads — it stays at the level of naming tools and parameters rather than showing complete invocation examples. The filter syntax examples are concrete and helpful. | 4 / 5 |
Workflow Clarity | Workflows are clearly sequenced with labeled steps and tool names, and steps are tagged as [Required], [Optional], [Prerequisite]. However, for destructive operations (DELETE_TASK, DELETE_SECTION), there are no validation or confirmation checkpoints. The bulk create workflow also lacks any verification step to confirm tasks were created successfully. Per the rubric, missing validation for destructive/batch operations caps this at 3. | 3 / 5 |
Progressive Disclosure | The content is well-structured with clear section headers and a useful quick reference table, but at ~250 lines it's quite long and monolithic. The repeated information across sections (pitfalls, due date handling, ID formats) could be consolidated or split into reference files. There are no bundle files or external references to offload detailed content like the full parameter reference or pitfalls catalog. | 3 / 5 |
Total | 13 / 20 Passed |