BC twzrd-trust
Discover x402 callables then check the seller BEFORE paying. Free resource join (GET /v1/intel/resources) lists listed|live_402 claims; free preflight returns a ReadinessCard (allow / warn / block) from the observed Solana x402 corpus. Composes with any x402 payer skill: discover → merchant_card wash refuse → preflight → gate_eval (AutoGate) when you control signing → optional pay; abort on decision=block. WHAT YOU GET FREE: resource join (source of truth), wash overlay on ingested listings (PayAI not_indexed), pre-spend ReadinessCard, merchant_card (wash_flagged refuse), wallet scores, secondary payer leaderboard research, counterparty + facilitator footprint, wash/sybil detection, batch + compare, offline receipt verify; route settle through TWZRD for free merchant_attach + twzrd_receipt on POST /settle. PAID (x402, USDC on Solana): full trust model + V6 receipt at GET /v1/intel/trust/{pubkey} (0.05 USDC); merchant track-record at GET /v1/intel/merchant/{pay_to} (0.05 USDC); score-only teaser at GET /v1/intel/quick/{pubkey} (0.001 USDC). TRIGGERS: should I pay this, is this wallet safe, check seller, x402 preflight, scam check, counterparty risk, wallet reputation, trust score, verify receipt, before paying, solana wallet check, agent trust, readiness card, wash flagged, merchant card, resource join, discover x402, facilitator settle, merchant attach, track record
Discover x402 callables then check the seller BEFORE paying.
As a process C 57/100 · Has gaps — weak spots: result and completion, inputs and preconditions, running it twice
How to improve
- Shorten the description to 1024 characters.
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 0. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 1390 chars, limit 1024 - warning
body-longSKILL.md body ≈ 5907 tokens (recommended < 5000); move details to references/ - note
frontmatter-keyunknown frontmatter key "homepage"
Process rating: all ten parameters 57/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 30Running it twice. 7 mutating operations with no state check
- 50When it triggers. No condition that starts the skill
- 60Tools and files. Uses tools (bash, web, python, node) that frontmatter does not declare
- 70Execution cost. Instruction body is 5907 tokens
- 100Steps. 26 steps
- 100Failures and branches. 2 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 12 top-level sections: this looks like several domains in one skill
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 1389: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +4Structure: 26 headings
- +3Step-by-step instructions: 26 items
- +4Has examples (20 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 45.