AB agentcash
Pay-per-call x402/MPP APIs (USDC on Base, Solana, Tempo). No API keys—wallet pays per request. If the task matches a SERVICES origin below, SKIP search and go straight to discover → fetch. Only search when NO listed origin fits. SERVICES: stableenrich (people/company, web search, scraping, Maps, LinkedIn, email verify, news), stablesocial (TikTok, Instagram, Facebook, Reddit, LinkedIn), stablestudio (AI image/video), stableupload (file/site hosting), stableemail (email, inboxes, subdomains), stablephone (AI calls, phone numbers), stablejobs (jobs), stabletravel (travel), stablebrowser (browser automation). TRIGGERS: research, enrich, scrape, search the web, generate image, video, social media, send email, phone call, travel, jobs, find contact, find API, x402, mpp, agentcash, register agent, accept payments, earn, pump.fun, token scoring
Pay-per-call x402/MPP APIs (USDC on Base, Solana, Tempo).
As a process B 70/100 · Nearly there — weak spots: inputs and preconditions, running it twice
How to improve
- 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 · 2
✓ No critical or high findings
Medium and low: 2
-
low Secrets in code
secret-high-entropy-tokenreferences/pumpfun-data-sources.md:41High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)--data '{"jsonrpc":"2.0","id":1,"method":"getSignaturesForAddress","params":["6EF8…F6P",{"limit":10}]}'detector -
low Secrets in code
secret-high-entropy-tokenreferences/x402-agent-pay.md:28High-entropy token-like string (may be an id, hash or a credential)0x36…a03
Files scanned: 4. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
frontmatter-keyunknown frontmatter key "homepage" - note
edit-residuethe text marks something as outdated (lines 73, 79): check that old rules are not kept next to new ones — the full check reads the text for contradictions
Process rating: all ten parameters 70/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 30Running it twice. 3 mutating operations with no state check
- 60Tools and files. Uses tools (bash, web) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 70When it triggers. States when to use, but not when not to
- 100Steps. 16 steps
- 100Failures and branches. 1 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 1989 tokens
- 100Progress reporting. Reports progress
- low The response is described with custom markup (6 tags): a typed call is more reliable
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 850: 120–800 characters recommended
- -42 reference files, but SKILL.md never points to them: the model will not open them
- +1No license
- +2Single-language instructions
- +4Structure: 18 headings
- +3Step-by-step instructions: 16 items
- +3Output format is stated explicitly
- +4Has examples (8 code blocks)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 79.