More screenshots
Works with the AI Agents you already use
SentryKit Lite
Offline secrets scanner and OWASP Top 10 security auditor for local code reviews.
Free
SentryKit Lite
Example session with this skill installed
Run a secrets scan on my current directory and triage the results. Mask any findings.
- Read your context and instructions
- Compiled the sentrykit lite
I have completed the offline secrets scan of the current directory. The scanner identified two potential issues which I have triaged based on context.
SentryKit Scan Results
| File | Finding | Type | Status | Triage Note |
|---|---|---|---|---|
config/services.yaml | sk_live_**** | Stripe API Key | Confirmed | Found in production config; needs immediate rotation. |
tests/auth_test.py | dckr_pat_**** | Docker Token | Dismissed | Verified as a dummy placeholder for unit testing. |
.env | OPENAI_API_KEY | Environment File | Confirmed | Sensitive file is not ignored by git. |
I have updated sentrykit-report/findings.json with these statuses.
Next steps
- Rotate the Stripe API Key immediately via the Stripe Dashboard; deleting the line is not sufficient.
- Add
.envto your.gitignorefile to prevent future accidental commits. - Run
sentrykit-report/report.mdto view the full impact assessment.
SentryKit assists review, not a replacement for a professional audit or pentest. Findings are heuristic and may include false positives or miss issues.
Connects securely to your tools. The creator never sees your data.
About this skill
Hardcoded secrets and common web vulnerabilities are the leading causes of preventable breaches. SentryKit Lite provides an offline security review starter that identifies leaked credentials and performs a targeted OWASP Top 10 audit directly within your development environment. It focuses on high-impact areas like authentication bypasses and injection flaws without requiring external services.
What it does
- Secrets scanning detects AWS, GitHub, Stripe, and OpenAI keys, plus JWTs and PEM files.
- OWASP Top 10 review audits entry points for access control, injection, and crypto failures.
- Local-only execution ensures code and credentials never leave your machine or hit the network.
- Safe reporting masks discovered secrets (e.g.,
AKIA****) and generates structured JSON findings. - Diff-mode analysis scans only changed lines to catch security regressions before a commit or PR.
How it works
- Run the
scan_secrets.pyscript against a local directory or specific git diff. - Triage findings in
findings.jsonto confirm real leaks versus test fixtures or placeholders. - Perform a guided audit of application entry points against the
owasp-top10-2025.mdreference. - Generate a final
report.mdincluding CWE identifiers, exploit scenarios, and suggested patches.
Frameworks & tools
This tool uses a Python 3 standard library scanner. It is compatible with any repository containing secrets or web entry points, including Node.js, Python, Go, and PHP applications.
Why this beats prompting it yourself
Generic prompts often miss low-entropy secrets or hallucinate security standards. This skill provides a structured methodology and a specialized local scanner that enforces masking, ensuring you don't accidentally leak full credentials into your LLM history.
Use cases
- Scan a repository for leaked API keys before moving a private project to public.
- Audit a new Pull Request for IDOR vulnerabilities and SQL injection points.
- Verify that
.envfiles or high-entropy strings haven't been committed to history. - Conduct a fast, offline security baseline check for a new microservice.
Known limitations
Requires Python 3 installed locally. It cannot detect vulnerabilities in compiled binaries or perform deep reachability analysis of third-party library internals.
Requires local Python 3. Cannot scan compiled binaries or perform deep reachability analysis. Read-only on source code; fixes are provided as separate patch files.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 2 days ago
- Free with an Agensi account
Requires