More screenshots
Works with the AI Agents you already use
LLM App Security Triage
An application that passes user text to a model has an attack surface that no traditional scanner covers.
$10
LLM App Security Triage
Example session with this skill installed
We added a support assistant that reads user tickets and can look up orders through a tool call. Before we ship, triage the llm-assistant/ folder for the LLM-specific problems and tell me where to look first.
- Read your context and instructions
- Compiled the llm app security
LLM App Security Triage Report
Scanner: llm-app-triage v1.0.0
Target: llm-assistant (working tree)
Files scanned: 10 | Findings: 21 | Files with findings: 6
Highest severity: critical | Rules fired: 20 of 20 | Families: 6 of 6
The scan read every text file it could parse and applied all 20 rules in the catalogue. 21 findings landed in 6 files. The other 4 files were silent, which is the point: the scanner is not tuned to shout at ordinary code.
Findings
| Rule | Family | Severity | File | Line | What it matched |
|---|---|---|---|---|---|
MOD-001 | model output execution | critical | src/planner.py | 7 | Model output evaluated as code |
PI-001 | prompt injection | critical | docs/support-ticket.txt | 3 | Text that tries to discard earlier instructions |
PI-002 | prompt injection | critical | docs/support-ticket.txt | 5 | Text that tries to hide activity from the operator |
SEC-001 | secrets | critical | config/settings.py | 3 | Cloud access key identifier |
SEC-002 | secrets | critical | config/settings.py | 5 | Source-hosting access token |
SEC-003 | secrets | critical | config/settings.py | 7 | Messaging platform token |
SHL-001 | shell construction | critical | src/agent_tools.py | 24 | Shell interpreter flag enabled on a call |
SHL-002 | shell construction | critical | src/agent_tools.py | 16 | Command handed to an interpreter call built from parts |
MOD-002 | model output execution | high | src/planner.py | 11 | Model output handed to a process call |
PI-003 | prompt injection | high | docs/support-ticket.txt | 7 | Text that tries to extract the model instructions |
PI-004 | prompt injection | high | docs/support-ticket.txt | 9 | Text that tries to put the model into an unrestricted mode |
SEC-004 | secrets | high | config/settings.py | 9 | Credential assigned to a literal value |
SHL-003 | shell construction | high | src/agent_tools.py | 12 | Command built with an interpolated value |
SHL-004 | shell construction | high | src/agent_tools.py | 20 | Interpreter call given a bare variable |
SQL-001 | query construction | high | src/reporting.py | 7 | Query keyword concatenated with a value |
SQL-002 | query construction | high | src/reporting.py | 11 | Query keyword built with an interpolated value |
SQL-003 | query construction | high | src/reporting.py | 15 | Query keyword built with percent formatting |
TAI-001 | untrusted data in the prompt | high | src/retrieval.py | 12 | Untrusted-looking value concatenated into a prompt |
TAI-002 | untrusted data in the prompt | medium | src/retrieval.py | 17 | Prompt built with an interpolated value |
TAI-003 | untrusted data in the prompt | medium | src/retrieval.py | 7 | Template placeholder filled from user input |
TAI-003 | untrusted data in the prompt | medium | src/retrieval.py | 22 | Template placeholder filled from user input |
Families that fired
| Family | Rules | Findings | Highest severity |
|---|---|---|---|
| prompt injection | PI-001, PI-002, PI-003, PI-004 | 4 | critical |
| secrets | SEC-001, SEC-002, SEC-003, SEC-004 | 4 | critical |
| shell construction | SHL-001, SHL-002, SHL-003, SHL-004 | 4 | critical |
| query construction | SQL-001, SQL-002, SQL-003 | 3 | high |
| untrusted data in the prompt | TAI-001, TAI-002, TAI-003 | 4 | high |
| model output execution | MOD-001, MOD-002 | 2 | critical |
Severity breakdown
| Severity | Findings |
|---|---|
| critical | 8 |
| high | 10 |
| medium | 3 |
How the evidence is shown
A credential finding never echoes the value. The scanner reports credential-shaped value on this line and points at the file and the line, so the report can be pasted into a ticket without leaking the key a second time. An injection finding reports instruction phrasing detected for the same reason: quoting the hostile phrase back would copy it into your ticket, your log and your chat history.
The shell, query, taint and model families quote the source line, because in those families the construct itself is the finding and it carries no key material.
Recommendation
Start with the two model-output findings and the four injection findings, because those six are the ones where text changes behaviour instead of merely looking wrong. MOD-001 is the worst line in the tree: a model response handed straight to an interpreter turns a hallucination into code execution. PI-001 and PI-002 sit in a support ticket, which means the hostile text arrived through the retrieval path and would have been inside the context window. The four secret findings in config/settings.py are rotation items rather than cleanup items, because a key that reached the remote is already compromised.
The shell and query findings are the ordinary part of an LLM backend and they are still real: a command assembled from parts and a query assembled by concatenation are code execution and injection waiting for a reachable value. The three taint findings are the structural defect underneath all of it, because user input is concatenated straight into a prompt with nothing separating it from the instructions.
Every finding carries a rule ID, a file and a line, so each one is a ticket rather than a paragraph, and the rule identifiers are stable, so the next scan can be compared against this one.
Next steps
- Remove the
evalon the model response and parse it against a fixed schema instead. - Treat every document that reaches the model as untrusted input: keep it in a delimited block and say that it is data.
- Rotate the four credentials at their issuers, then load them from the environment or a secret manager at run time.
- Replace the concatenated and formatted queries with parameterised statements.
- Rebuild the command calls as argument lists and turn the shell interpreter flag off.
- Separate user input from the instructions in the prompt builder, then re-run the triage and confirm the findings are gone.
Connects securely to your tools. The creator never sees your data.
About this skill
An application that passes user text to a model has an attack surface that no traditional scanner covers. The request can try to redirect the model, the retrieval index can contain poisoned documents, a tool call can be steered to an unsafe operation, and the model's answer can be rendered back to a browser without escaping.
This skill triages an LLM-backed application against twenty rules in six families: prompt injection, secrets, shell construction, query construction, untrusted data reaching the prompt, and model output reaching an interpreter. It works over source files, over logs, and over a single sample of model input, which makes it usable in three different places: before a release, during an incident, and when you want to understand one captured prompt.
The analyser is pattern based and deterministic. It does not call a model, it does not need an API key, and it does not send anything anywhere. It is a first pass that tells you where to look, with stable rule identifiers so two runs can be compared.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Security.
The analyser reads one file at a time, so it cannot follow a value from one file to another - that trace is a manual step. Pattern matching catches known shapes, not novel attacks. It does not call a model or your provider API, and it does not execute the code under review.
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 buying a skill.
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
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 1 day ago
- Software
Python 3.9 or newer
Permissions
File Scopes
Shell is used to run the bundled analyser and read access is used to load the files you point it at. It writes no files, opens no network connections and reads no environment variables.