Content
57%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 content is genuinely actionable — nearly every section is anchored by a copy-paste-ready membrane CLI command, and the connection state machine is well documented. Its main weaknesses are broken cross-references ("Step 2" never exists), padding that assumes less competence than needed (intro and entity list), and a monolithic single-file structure that inlines generic Membrane CLI reference material.
Suggestions
Fix the workflow sequence: label the steps explicitly (1a/1b/2) so the two references to "skip to Step 2" point to a section that actually exists, likely "Searching for actions".
Delete the marketing-style intro paragraph and the "Amity Overview" bullet tree (or replace it with a mapping of Amity entities to concrete actions), and trim the rationale sentences in Best Practices.
Move the generic Membrane CLI material (connection states, proxy flag table, agent types) into a references/ file and keep SKILL.md as a lean Amity-specific overview with one-level-deep pointers.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly command-focused, but the three-sentence marketing intro ("Amity is a social platform that provides pre-built features..."), the "Amity Overview" bullet tree that lists eight entities with no actionable use, and rationale sentences like "This will burn less tokens and make communication more secure" are unnecessary padding that could be trimmed, fitting anchor 3 rather than anchor 4's minor instances. | 3 / 5 |
Actionability | Concrete, executable commands throughout (npm install, membrane login, connection ensure with --wait polling, action list/run with --input, a proxy flags table), but all examples use placeholders — no actual Amity action id or real API endpoint path (e.g. a /api/v5 path) is ever shown, so it falls just short of anchor 5's fully-covered common cases. | 4 / 5 |
Workflow Clarity | A sequence exists (install, login, connect, poll states, search, run) with decent state-checkpoint handling for CLIENT_ACTION_REQUIRED, but the document twice says "skip to Step 2" and has a "#### 1b" heading with no Step 2 or 1a anywhere, making the sequence partially incoherent and fitting anchor 3's gaps. | 3 / 5 |
Progressive Disclosure | This is a single-file skill over 150 lines with no references/ or scripts/ bundle and all content inline; roughly half the body is generic Membrane CLI usage (connection states, proxy flags, agent types) that clearly belongs in a shared or separate reference file, fitting anchor 3's "content that should be separate is inline" rather than anchor 4's mostly-appropriate placement. | 3 / 5 |
Total | 13 / 20 Passed |