AC php
Writes, debugs, and reviews PHP: type juggling, arrays, OOP, Composer, PDO, sessions, PHP-FPM, OPcache, PHP 8 features. Use when PHP returns a blank white page or a 500 with no message, when "headers already sent", "Allowed memory size exhausted", "Maximum execution time exceeded", "Class not found" after a Composer install, or an undefined array key warning appears; when `==` compares wrong, `strpos` or `preg_match` returns a falsy match, accents turn into mojibake, `DateTime` shifts by a day or an hour, `json_encode` returns false, PDO refuses to bind a parameter or emulates prepares, sessions serialize parallel AJAX requests, nginx answers 502 or 504 in front of PHP-FPM, OPcache keeps serving old code, or a version upgrade breaks on deprecations. Also for escaping output, hashing passwords, sizing FPM workers, writing PHPUnit tests, and clearing PHPStan errors. Not for framework internals — `laravel` covers Eloquent, queues, and Laravel auth.
As a process C 64/100 · Has gaps — weak spots: inputs and preconditions, running it twice, progress reporting
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 · 0
✓ No critical or high findings
Files scanned: 26. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
frontmatter-keyunknown frontmatter key "slug" - note
frontmatter-keyunknown frontmatter key "changelog" - note
frontmatter-keyunknown frontmatter key "homepage"
Process rating: all ten parameters 64/100
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Progress reporting. Says nothing while it works
- 30Running it twice. 11 mutating operations with no state check
- 50When it triggers. No condition that starts the skill
- 60Tools and files. Uses tools (bash) that frontmatter does not declare
- 60Result and completion. Output format stated, no completion criterion
- 70Execution cost. Instruction body is 4745 tokens
- 100Steps. 44 steps
- 100Failures and branches. 3 branches, has a failure section
- 100Consistency. Name and required fields are in place
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low 10 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
- +3Description length 959: 120–800 characters recommended
- +4No input/output examples
- +1No license
- +2Single-language instructions
- +5Description quotes 4 example trigger phrases
- +4Description says when NOT to use the skill
- +4Structure: 10 headings
- +3Step-by-step instructions: 44 items
- +3Output format is stated explicitly
Quality base 70; lint remarks subtract, signals add up to 100. Result: 86.