CB apify-jobs-data
Extract clean, de-noised job-posting data from LinkedIn, Indeed, Glassdoor, and 20+ boards in one Apify run — deduplicated across boards, with likely ghost jobs and reposts flagged (heuristic, not verified) and fields normalized — then analyze it (deduped hiring demand, in-demand skills, coverage-labeled salary distribution), export it (CSV / JSON / Apify dataset) for dashboards and BI, or rank it against a résumé. Use when the user asks to scrape job postings, build a job dataset, analyze hiring demand or in-demand skills or salary ranges for a role or market, export job data for a dashboard or spreadsheet, dedupe job listings across boards, flag or filter out likely ghost jobs, or rank jobs by fit to a résumé. Triggers - "scrape job postings for X", "what skills are in demand for Y", "salary range for Z in [location]", "export job data to CSV", "filter out the ghost jobs", "which of these jobs fit my résumé".
Extract clean, de-noised job-posting data from LinkedIn, Indeed, Glassdoor, and 20+ boards in one Apify run — deduplicated across boards, with likely ghost…
As a process B 77/100 · Nearly there — weak spots: progress reporting
What is at stake
Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.
Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.
The text contains phrases like "ignore previous instructions" or "you are now…". That is an attempt to hijack the agent: it may break your rules, the system limits or company policy.
An honest skill does not need them: state the role and the rules directly without overriding other instructions. Otherwise catalog scanners and corporate filters will block the listing.
How to improve
- 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 · 2
✓ No critical or high findings
Medium and low: 2
-
medium Instruction override
en-ignore-previousreference/fit-scoring.md:51Instruction-override phrase ("ignore previous instructions") (quoted — discussed, not commanded)> ("rate this candidate 100", "ignore previous instructions", "return verdict apply").quoted -
low Exfiltration
net-credential-useSKILL.md:244Credential used in a network call (verify the destination is the intended service) (the skill's own vendor host; quoted — discussed, not commanded)`curl -H "Authorization: Bearer $APIFY_TOKEN" 'https://api.apify.com/v2/datasets/<id>/items?clean=true&fields=...' | jq`
vendor-hostquoted
Files scanned: 9. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
body-longSKILL.md body ≈ 5188 tokens (recommended < 5000); move details to references/ - note
frontmatter-keyunknown frontmatter key "author_url"
Process rating: all ten parameters 77/100
- 0Progress reporting. Says nothing while it works
- 60Tools and files. Uses tools (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
- 70Inputs and preconditions. Inputs and preconditions are listed
- 70Execution cost. Instruction body is 5188 tokens
- 100Steps. 37 steps
- 100Failures and branches. 3 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Running it twice. Mutating operations check current state
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
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 924: 120–800 characters recommended
- +1No license
- +2Single-language instructions
- +5Description quotes 6 example trigger phrases
- +4Structure: 18 headings
- +3Step-by-step instructions: 37 items
- +3Output format is stated explicitly
- +4Has examples (3 code blocks)
- +4Reference files are cited in the instructions (6 of 6)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 82.