Inline ignore directives allow you to suppress Aguara findings on specific lines of code using special comments. This is useful for intentional patterns that would otherwise trigger false positives.Documentation Index
Fetch the complete documentation index at: https://mintlify.com/garagon/aguara/llms.txt
Use this file to discover all available pages before exploring further.
Basic Syntax
Place an inline comment containingaguara-ignore followed by one or more rule IDs:
CRED_004 (credential leak) on the same line is suppressed.
Comment Styles
Aguara recognizes multiple comment formats:#, //, --, <!--
Directive Types
Same-Line Ignore
Suppress a rule on the same line as the directive:Next-Line Ignore
Suppress a rule on the next line using-next-line:
Multiple Rules
Suppress multiple rules by separating IDs with commas:Ignore All Rules
Suppress all rules on a line by omitting the rule ID:Complete Examples
Test Fixtures
SKILL.md
Configuration Files
.aguara.yml
Documentation Examples
Intentional Downloads
How It Works
When Aguara scans a file:- Parse directives - Extract all
aguara-ignorecomments and build an index of which lines suppress which rules - Run analyzers - Pattern matching, NLP, taint tracking, and rug-pull detection run normally
- Filter findings - Before returning results, findings matching suppressed (line, rule ID) pairs are removed
- Scan output (terminal, JSON, SARIF, Markdown)
- Exit code calculation (for
--fail-on) - Finding counts
Inline ignores only affect findings on the exact line number specified. If a finding spans multiple lines, place the directive on the line where the match occurs (usually the first line).
Regex Pattern
Aguara uses this regex to detect inline ignores:- Optional comment prefix:
#,//,--,<!-- - Literal
aguara-ignore - Optional
-next-linesuffix - One or more uppercase rule IDs separated by commas
- Optional HTML closing
-->
- Lowercase rule IDs:
aguara-ignore cred_004❌ - No whitespace:
#aguara-ignoreCRED_004❌ - Inline with code:
api_key = 'foo' # aguara-ignore CRED_004❌ (must be on its own line or before code)
Limitations
Cannot Ignore by Category
You cannot suppress all rules in a category:aguara-ignore (all rules).
Cannot Ignore Ranges
There is no “start ignore / end ignore” block syntax. Each line must be suppressed individually:Does Not Affect Rug-Pull Detection
Inline ignores suppress findings from pattern matching, NLP, and taint tracking, but not rug-pull detection (RUGPULL_001). Rug-pull findings are based on file hash changes, not line-level patterns.
Best Practices
Always Include Rule IDs
Prefer specific rule IDs over blanket ignores:Add Context Comments
Explain why the pattern is safe:Use in Test Files
Test fixtures often contain intentional attack patterns:tests/fixtures/prompt-injection.md
Prefer Configuration Over Inline Ignores
If you need to suppress a rule across many files, use.aguara.yml:
.aguara.yml
Alternatives
Before using inline ignores, consider:- Rule overrides - Lower severity or disable rules project-wide in
.aguara.yml - Ignore patterns - Skip entire files or directories via
.aguaraignore - Code refactoring - Change the pattern to avoid triggering the rule
Verification
To verify an inline ignore worked:-
Before adding the directive, run the scan and note the finding:
-
Add the directive on or before the line:
-
Re-run the scan and confirm the finding is gone:
-
Check JSON output to confirm the finding was suppressed (not just hidden):
Related
- Configuration File - Project-wide settings
- Rule Overrides - Disable or change severity globally
- Ignore Patterns - Skip files and directories
