Works with the AI tools you already use
SAST Finding Triage Skill
Triage SAST security findings by deduplicating noise, validating context, and prioritizing remediation queues.
$5.99
SAST Finding Triage Skill
Example session with this skill installed
Triage these 3 findings from our recent SonarQube SARIF export. Include asset criticality 'High' for the 'billing-service' repository and identify if any are duplicates or false positives.
- Read your context and instructions
- Compiled the sast finding triage
- Generated the document
The SARIF export for the billing-service has been processed. I have normalized 3 findings into 2 unique issue groups, identified one likely false positive due to framework-level sanitization, and prioritized the remaining critical vulnerability.
SAST Triage Executive Summary
- Total findings: 3
- Unique findings/groups: 2
- Confirmed/likely issues: 1
- False-positive candidates: 1
- Highest-priority area: SQL Injection in
db_manager.py - Data limitations: Missing authentication context for session findings.
Finding Triage Table
| Finding/Group | Rule | Category | Severity | Confidence | Classification | Priority | Reason |
|---|---|---|---|---|---|---|---|
| G-001 | S2100 | SQLi | Critical | High | Confirmed | P0 | Direct user input to cursor.execute |
| F-002 | S5144 | SSRF | Medium | Low | Likely FP | P3 | Input passed through url_validator |
Remediation Queue
| Priority | Finding/Group | Recommended Action | Owner | Verification |
|---|---|---|---|---|
| P0 | G-001 (SQLi) | Implement parameterized queries via SQLAlchemy core | Platform Team | Re-scan and manual PR review |
| P3 | F-002 (SSRF) | Document existing validator as a compensating control | Security Team | Validation logic audit |
Developer Handoff
Issue: SQL Injection in billing-service/db_manager.py
Location: Line 42, fetch_user_record function.
Impact: Unauthorized database access and data exfiltration.
Remediation: Replace string formatting in the query builder with prepared statements. Ensure all inputs to the query are cast to expected types before execution.
Acceptance Criteria: The code must pass the SAST S2100 check and undergo a manual peer review of the database layer.
Next steps
- Export the G-001 handoff as a Jira ticket for the Platform Team.
- Update the suppression list for F-002 with the provided evidence of validation.
- Provide the full
db_manager.pysource if you need specific code refactoring for G-001.
sast-finding-triage-skill.pdf
PDF · document
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
What you get
About this skill
The problem
Security scanners generate hundreds of noisy alerts, often burying critical vulnerabilities under duplicates and false positives. Developers lose hours manually verifying scanner output instead of fixing code.
What it does
- Normalizes heterogeneous findings from SARIF, JSON, or CSV exports into a standard schema.
- Groups duplicates by root cause and data-flow similarity to reduce alert volume.
- Validates source-to-sink reachability and framework protections to identify false-positive candidates.
- Calculates remediation priority by weighing technical severity against asset criticality and exposure.
- Generates structured developer handoff tickets with actionable remediation guidance.
Frameworks & tools
Works with SARIF-compliant tools, OWASP/CWE mappings, and standard SAST exports. Compatible with major framework security patterns and CI/CD pipeline metadata.
Why this beats prompting it yourself
Generic prompts often conflate scanner severity with actual business risk or hallucinate data flow. This skill enforces a strict evidence-based triage logic that separates technical confidence from business priority and prevents the suppression of findings without documented justification.
Use cases
- Cleaning up a massive backlog of legacy security debt after a new scanner rollout.
- Pre-processing CI/CD security gate results to prevent breaking builds on known false positives.
- Converting raw security audits into prioritized Jira tickets for engineering teams.
Known limitations
Requires specific evidence to confirm exploitability. Missing metadata is marked as unknown rather than inferred.
How to install
Works the same in every agent - Claude, Cursor, Codex, Copilot and 20+ more.
- 1
Download the ZIP
Free skills download straight away. Paid skills unlock right after purchase.
- 2
Unzip into your skills folder
Every agent reads skills from one folder on your machine. Drop the unzipped folder in there.
- 3
Ask your agent to use it
Restart the agent if it was already running. It picks the skill up automatically - no config needed.
Skills folder by agent
Click the path to copy it. Create the folder if it does not exist yet.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
30-day refund
Not a fit? Get your money back
Trust & safety
Security scanned
Verified clean 3 days ago
- Passed all security checks, Safe to install