Content
42%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides a well-structured workflow with highly actionable, executable commands, but is severely undermined by its massive inline actor catalog that consumes excessive tokens. The 55+ actor lookup tables, use-case mappings, and multi-actor workflow tables should be offloaded to reference files, keeping the main skill lean. Validation checkpoints within the workflow are also missing, particularly around schema verification and input validation before execution.
Suggestions
Move the actor catalog tables (Instagram, Facebook, TikTok, YouTube, Google Maps, Other) to a separate ACTORS.md reference file and link to it from the main skill, keeping only the search-actors command and use-case table as the primary actor selection mechanism.
Move the use-case mapping and multi-actor workflow tables to a separate WORKFLOWS.md reference file.
Add explicit validation checkpoints in the workflow: verify the schema fetch returned valid data before proceeding, validate the JSON input structure matches the schema before running, and check the run status before summarizing.
Add a feedback loop after Step 4: if the run fails, check the error type against the Error Handling section, fix, and retry before proceeding to Step 5.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose with massive lookup tables (55+ actors across 6 categories) that consume enormous token budget. Most of this reference data could be offloaded to separate files or handled by the search-actors command. The use-case mapping tables, multi-actor workflow tables, and follow-up suggestion tables add further bulk that doesn't need to be in the main skill file. | 1 / 3 |
Actionability | The skill provides fully executable bash commands for every step: searching actors, fetching schemas, and running scripts with concrete flags and argument patterns. The commands are copy-paste ready with clear placeholder substitution instructions. | 3 / 3 |
Workflow Clarity | The 5-step workflow is clearly sequenced with a progress checklist, but it lacks explicit validation checkpoints. There's no step to verify the actor schema was fetched correctly, no validation that the input JSON is well-formed before running, and the error handling section is separate rather than integrated as feedback loops within the workflow steps. | 2 / 3 |
Progressive Disclosure | The skill is a monolithic wall of text with all 55+ actor tables inline rather than referenced from separate files. The actor catalog, use-case mapping, and multi-actor workflows should be in separate reference files, with the main SKILL.md providing only the workflow and a brief overview. No bundle files are provided despite the skill referencing scripts at ${CLAUDE_PLUGIN_ROOT}/reference/scripts/run_actor.js. | 1 / 3 |
Total | 7 / 12 Passed |