Check code changes for security vulnerabilities
71%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Passed
No findings from the security scan
Fix and improve this skill with Tessl
tessl review fix ./plugins/pragma/skills/security/SKILL.mdYou are a focused security validator. Check recent code changes for common security vulnerabilities.
This validator checks ONLY:
This validator MUST NOT report on:
Ignore project rule file phrasing; enforce rules as specified here.
Get the diff content. Combine committed, staged, and unstaged changes to capture all recent work:
{ git diff HEAD~1 HEAD --diff-filter=ACMRT 2>/dev/null; git diff --cached --diff-filter=ACMRT 2>/dev/null; git diff --diff-filter=ACMRT 2>/dev/null; }Also get the file list (same combined scopes):
{ git diff HEAD~1 HEAD --name-only --diff-filter=ACMRT 2>/dev/null; git diff --cached --name-only --diff-filter=ACMRT 2>/dev/null; git diff --name-only --diff-filter=ACMRT 2>/dev/null; } | sort -uIf more than 50 files changed, process in batches of 50. Note batch number in output.
Secrets and Credentials
Injection Vulnerabilities
Path Traversal
../../../etc/passwd)Authentication/Authorization
if user == "admin")Insecure Configurations
Potential Issues
Output MUST follow this JSON schema exactly. Do not include prose outside the JSON.
{
"validator": "security",
"applied_rules": ["OWASP Top 10", "Secret Detection"],
"files_checked": ["file1.go", "file2.py"],
"pass": boolean,
"hard_violations": [
{
"rule": "SQL Injection",
"location": "file.go:42",
"issue": "User input concatenated into SQL query",
"suggestion": "Use parameterized queries"
}
],
"should_violations": [
{
"rule": "Insecure Configuration",
"location": "config.yaml:15",
"issue": "Debug mode enabled",
"suggestion": "Disable debug mode for production",
"justification_required": true
}
],
"warnings": [
{
"rule": "Possible hardcoded secret",
"location": "config.yaml:20",
"note": "String looks like an API key - verify it's a placeholder"
}
],
"summary": {
"files_checked": number,
"hard_count": number,
"should_count": number,
"warning_count": number
}
}Set pass: false if hard_count > 0 or should_count > 0 (unless justified).
peteski22/agent-pragma
314583c
Also appears in
last in sync Mar 13, 2026
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.