CC livecheck
Primary use: clean a batch of job URLs before apply or outreach. Live status of a specific job posting, product page, or listing, read from the page itself right now, via x402. $0.01 per posting, no account or API key. Returns live / closed / unknown plus title and signals (in-stock, sold-out, apply form present, http_404). Also one-shot condition checks and 30-day URL watchers. USE FOR: - Cleaning a batch of job URLs before apply or outreach (ghost jobs, stale board rows, scraped lists) - Click-time check that a job posting is still open before tailoring a resume, applying, or spending the user's credits - The same live / closed / unknown check across Greenhouse, Lever, Workday, Ashby, SmartRecruiters, and iCIMS - Checking whether a product page, eBay item, or Shopify listing is still available before recommending or buying it - Confirming a search result (Google Shopping, job board, scraper output) isn't stale - Checking whether a price crossed a threshold or a keyword appeared on a page, once - Watching a URL for 30 days and getting a signed webhook when it changes TRIGGERS: - "clean this list", "which of these jobs are still open", "before I reach out", "ghost job" - "is this job still open", "position filled", "still accepting applications", "sold out" - "is this still available", "still live", "still in stock" - "posting closed", "still hiring" - "dead link", "404", "check these URLs", "which of these are live" - "before I apply", "before I buy", "before scraping" - "watch this page", "tell me when", "price drops below", "back in stock" Use `npx agentcash@latest fetch` for livecheck.fly.dev endpoints. Job and listing checks are $0.01 per URL; generic verify is $0.01; one-shot checks $0.02; watchers $2.50 for 30 days. On HTTP 503, wait for the Retry-After header (seconds) and retry. For a list of URLs, send up to 8 verifies at a time (server concurrency default is 8; a short queue may absorb brief bursts).
Primary use: clean a batch of job URLs before apply or outreach.
As a process C 55/100 · Has gaps — weak spots: result and completion, when it triggers, inputs and preconditions
How to improve
- Shorten the description to 1024 characters.
- 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: 3. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- error
description-longdescription is 1949 chars, limit 1024 - note
description-budgetdescription takes 1949 of the ~15000-char shared budget for all skills
Process rating: all ten parameters 55/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 20When it triggers. No condition that starts the skill
- 30Running it twice. 11 mutating operations with no state check
- 60Tools and files. Uses tools (bash, web) that frontmatter does not declare
- 100Steps. 35 steps
- 100Failures and branches. 1 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 3145 tokens
- 100Progress reporting. Reports progress
- 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
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 1948: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +5Description quotes 23 example trigger phrases
- +4Structure: 16 headings
- +3Step-by-step instructions: 35 items
- +4Has examples (8 code blocks)
- +4Reference files are cited in the instructions (1 of 1)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 62.