Name: mercury
Rating: 55.2 (1 reviews)
Author: membranedev

Blog Docs Log in Get started

mercury

Mercury integration. Manage Organizations. Use when the user wants to interact with Mercury data.

Quality

62%

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/mercury/SKILL.md

Quality

Content

67%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 content is a strong, executable CLI integration guide with a clear connection/state-feedback workflow and well-organized sections. Its main flaws are non-authoritative noise — a leaked model-refusal line, a mismatched Postlight web-parser docs link, and an overview listing (Email/Draft/Contact/Label) that does not match the Mercury banking actions table.

Suggestions

Remove the leaked refusal line ('I don't have enough information to do that...') and the mismatched 'Official docs: https://mercury.postlight.com/web-parser/' link, which refers to a different Mercury product.

Fix the 'Mercury Overview' section so it reflects the actual banking actions (accounts, transactions, customers, invoices, recipients, treasury, credit) instead of the unrelated Email/Draft/Contact/Label list.

Add one fully worked example, e.g. `membrane action run list-accounts --connectionId=$CID --json`, and clean up the step numbering (label Step 1/Step 2 explicitly) so the 'skip to Step 2' reference resolves.

Dimension	Reasoning	Score
Conciseness	The body is mostly an efficient command reference, but it includes a leaked refusal line ('I don't have enough information to do that...'), a redundant 'Mercury Overview' that doesn't match the actions table, and filler like 'Use action names and parameters as needed.', matching 'Mostly efficient but includes some unnecessary explanation'. It is not a 4 due to those notable unnecessary bits, and above a 2 because the core commands are not padded.	3 / 5
Actionability	Concrete, copy-paste-ready commands are given throughout (install, login, connection ensure/get with --wait, action list/run, request proxy) plus a popular-actions table and a proxy-flags table, matching 'Mostly executable guidance; concrete code or commands with minor gaps'. It is not a 5 because no fully worked action-run example ties the generic pattern to a real action from the table, and the leaked refusal line adds noise.	4 / 5
Workflow Clarity	The install -> authenticate -> connect -> poll-for-ready -> search -> run sequence is clear and includes a genuine feedback loop (poll state, handle CLIENT_ACTION_REQUIRED/CONFIGURATION_ERROR, re-poll), matching 'Clear sequence with most checkpoints present; minor validation gaps'. It is not a 5 because step numbering is incoherent ('skip to Step 2' with no labeled Step 2, and '#### 1b' without a 1a).	4 / 5
Progressive Disclosure	The single-file body is organized into clear sections (Install, Authentication, Connecting, Searching, Popular actions, Running, Proxy, Best practices) with no bundle files present and no external references needed for this scope, matching 'Good structure; most content appropriately placed; minor organization gaps'. It is not a 5 because the redundant overview and leaked line are minor organization issues.	4 / 5
	Total	15 / 20 Passed

Description

57%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 has an explicit 'Use when' trigger and names the target brand, giving it reasonable completeness and distinctiveness. It is weakest on specificity, offering only one generic action ('Manage Organizations') rather than a concrete inventory of Mercury capabilities.

Suggestions

Replace the generic 'Manage Organizations' with concrete actions drawn from the skill, e.g. 'List accounts, transactions, customers, and invoices; create customers, invoices, and payment recipients.'

Add natural trigger phrases and synonyms users would say, such as 'Mercury bank', 'Mercury accounts', or 'Mercury transactions', to improve trigger-term coverage.

Tighten the 'when' clause to reference concrete Mercury banking scenarios (e.g. 'Use when the user wants to query bank accounts, transactions, or accounts-receivable data in Mercury').

Dimension	Reasoning	Score
Specificity	The description names the domain ('Mercury integration') but the only action, 'Manage Organizations', is minimal and generic, and 'interact with Mercury data' is abstract — matching the anchor 'Names the domain but actions are minimal or generic'. It falls short of a 3, which expects 1-2 more concrete actions, and is well above a 1 since the domain is explicitly named.	2 / 5
Completeness	Both 'what' ('Mercury integration. Manage Organizations.') and an explicit 'when' ('Use when the user wants to interact with Mercury data.') are present, matching 'Has both what and when; when could be more explicit or specific'. It is not a 5 because the 'what' is vague and lacks concrete trigger phrases, and not a 3 because the explicit 'Use when' clause satisfies the trigger-guidance requirement.	4 / 5
Trigger Term Quality	It includes the core natural keyword 'Mercury' plus 'Organizations' and 'data', but lacks common variations or synonyms, matching 'Some relevant keywords but missing common variations'. It is above a 2 (which expects only generic keywords) but below a 4 because the natural-phrase coverage is thin.	3 / 5
Distinctiveness Conflict Risk	Naming the specific brand 'Mercury' creates a clear niche and low conflict risk for users referencing the app, matching 'Mostly distinct; minor overlap risk'. It is not a 5 because the description is thin and 'Mercury' is a common word (element, planet, car brand) with minor overlap risk.	4 / 5
	Total	13 / 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/mercury/SKILL.md
Commit: f484c82

Reviewed: 6 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.