Public leaderboard

Public assessment

jacksun911/megalaunch-mcp (megalaunch-mcp)

megalaunch-mcp · v1.0.0 · scanned

What changed in the harness

Selection accuracy 98→100, token cost up 13%, unconfirmed writes 0%→0%.

Category breakdown

Where the score comes from.

Earned points across the four signals Gradable measures. Safety and Legibility are scored out of 30; Economics and Discoverability are scored out of 20.

01Safety

0.0 / 30

0.0 out of 30
02Legibility

27.4 / 30

27.4 out of 30
03Economics

20.0 / 20

20.0 out of 20
04Discoverability

12.4 / 20

12.4 out of 20

Highest-impact fix

Estimated gain +30 points

Add explicit identity and permission preflight tools

Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions.

Description evidence

Defects and rewrites.

6 defects found across the exposed tool descriptions. Suggested rewrites make purpose, inputs, boundaries, and returns easier for an agent to understand.

Tool Defect types Suggested rewrite
megalaunch_status
no_return_description
Check MegaLaunch service status and recent launch statistics. Returns whether the service is currently operational along with recent launch statistics, such as total launches and success rate.
megalaunch_pricing
no_return_description
Get current pricing for token launch packages (Basic and Premium). Returns the price and included features for each package.
megalaunch_create_order
params_unexplained
Create a new meme token launch order on Solana/pump.fun. Returns the deposit address and amount to send for payment. The package selects the launch tier (basic or premium); tokenName, tokenSymbol, and tokenDescription customize the token details shown to buyers; set aiPick to true to let the AI choose a trending name and symbol, optionally based on a theme (war, money, hype, animal, fire, ice).
megalaunch_list_orders
no_return_description
List your token launch orders with an optional status filter. Returns a list of your orders; use limit and offset to page through the results, and status to only return orders in that state.
megalaunch_get_order
no_return_description
Get the detailed status of a specific token launch order by its orderId. Returns the order's status and details, such as token information and payment state.
megalaunch_cancel_order
no_return_description
Cancel a pending token launch order by its orderId. Works only for unpaid orders. Returns confirmation that the order was cancelled, or an error if the order is already paid and cannot be cancelled.

Selection evidence

Confusable tool pairs.

4 pairs where similar names or overlapping descriptions may send an agent toward the wrong tool.

Tool A Tool B Confidence Why they collide
megalaunch_list_orders megalaunch_get_order medium Both retrieves order info; a vague task like 'show me my launch order' or 'find my order status' could send the agent to list all orders with a status filter instead of looking up one specific order, or vice versa, since neither description clearly disambiguates aggregated vs single-order queries.
megalaunch_status megalaunch_list_orders medium The status tool is literally named 'status' and mentions 'recent launch statistics', so a task like 'check the status of my token launch orders' pulls the agent toward status even though the user's own order status requires list_orders (or get_order); the name/description shadowing makes the wrong pick plausible.
megalaunch_pricing megalaunch_create_order low A task like 'how much for a basic token launch and get it going' is ambiguous between querying pricing and actually creating the order; an agent focused on the cost keyword picks pricing while the real intent was creation, though the verb 'create' vs 'pricing' mostly separates them.
megalaunch_create_order megalaunch_get_order low Sloppy phrasing such as 'get me an order started' or 'set up my launch and then check it' could route to either create or get since both share the order/get tokens, though the descriptions (creation inputs vs order ID lookup) usually resolve it.

Compare the field

One score is useful.
The evidence makes it actionable.

Back to the leaderboard