Name: sonix
Rating: 60.8 (1 reviews)
Author: membranedev

sonix

Sonix integration. Manage Deals, Persons, Organizations, Leads, Projects, Activities and more. Use when the user wants to interact with Sonix data.

Quality

71%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/sonix/SKILL.md

Quality

Content

75%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable with executable CLI commands and a clear connection state machine, scoring top marks on actionability. It loses points on conciseness from redundant command restatements and background prose, and on workflow clarity from the absence of validation/retry loops around action execution.

Suggestions

De-duplicate the action-discovery command, which appears in both 'Searching for actions' and 'Popular actions' — keep one canonical example and cross-reference it.

Trim the marketing-style overview sentence about who uses Sonix; Claude does not need audience framing to operate the integration.

Add a brief validate/retry checkpoint after 'membrane action run' (e.g. check the response output/error field and re-run or fall back to the proxy on failure) to close the workflow-clarity gap.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient with runnable commands, but it includes explanation Claude already knows (e.g. what Sonix is and who uses it) and restates the action-list command twice in 'Searching for actions' and 'Popular actions', which could be tightened.	3 / 5
Actionability	Commands are concrete and copy-paste ready (e.g. 'membrane connection ensure "https://sonix.ai/" --json', 'membrane action run <actionId> --connectionId=CONNECTION_ID --input ...'), and the proxy flag table plus JSON examples fully cover the common cases.	5 / 5
Workflow Clarity	The connection setup is a clear state-machine sequence with polling and explicit next-step branches (READY / CLIENT_ACTION_REQUIRED / CONFIGURATION_ERROR), but there is no validate-after-run or error-retry loop for the actions/proxy operations that follow.	4 / 5
Progressive Disclosure	Content is organized into well-signaled sections (Overview, Working with Sonix, Authentication, Connecting, Searching, Popular actions, Best practices) with no nested file references; the only gap is that some referenceable detail (the clientAction schema, proxy options) is fully inlined rather than split out.	4 / 5
	Total	16 / 20 Passed

Description

67%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description cleanly pairs a 'what' (manage named Sonix entities) with a 'when' trigger and uses third person, landing mostly in the upper anchors. Its weakness is that the trigger language is generic ('interact with Sonix data') and oddly lists CRM-style entities that do not match Sonix's transcription/subtitling domain, slightly capping trigger quality.

Suggestions

Replace the generic 'interact with Sonix data' trigger with domain-natural verbs, e.g. 'Use when the user wants to transcribe audio or video, generate subtitles, or translate transcripts in Sonix.'

Verify the entity list ('Deals, Persons, Organizations, Leads, Projects, Activities') against Sonix's actual API — these read as CRM objects and may be templated from another skill; align them to Sonix resources (files, transcripts, subtitles, workspaces, speakers).

Drop the 'and more' hedge in favor of one or two more concrete actions to push specificity from 4 toward 5.

Dimension	Reasoning	Score
Specificity	Lists several concrete managed entity types ('Deals, Persons, Organizations, Leads, Projects, Activities') rather than vague abstractions, though the 'and more' hedge leaves minor gaps in coverage.	4 / 5
Completeness	Explicitly answers both 'what' (manage Deals, Persons, Organizations, etc.) and 'when' ('Use when the user wants to interact with Sonix data'); the 'when' clause is present and concrete but could name specific Sonix tasks more precisely.	4 / 5
Trigger Term Quality	The trigger phrase 'when the user wants to interact with Sonix data' uses the brand name but offers only generic natural-language terms ('interact with data'); it omits the natural verbs users would actually say such as 'transcribe', 'translate', or 'subtitle'.	3 / 5
Distinctiveness Conflict Risk	The 'Sonix' brand trigger plus enumerated entities form a distinct niche with minimal overlap risk against other skills, though the generic 'interact with data' phrasing is slightly less crisp than a domain-specific trigger.	4 / 5
	Total	15 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: membranedev/application-skills
Path: skills/sonix/SKILL.md
Commit: f484c82

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.