Security audit

NDIS Incident Reporting Specialist

Security checks for vulnerabilities and agentic risk

Overview

This skill is a plain Markdown guidance-and-drafting aid for NDIS incident reports, with no hidden execution, persistence, or external data access found.

Before installing, treat this as a regulated-compliance drafting aid rather than legal advice. Use participant initials or IDs in drafts where possible, verify current NDIS Commission requirements before submission, and have ambiguous or serious incidents reviewed by a compliance lead or the Commission.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (8)

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The trigger description is unusually broad and includes generic phrases like "something happened with a client/participant" and "help me write an incident report," which can cause the skill to activate on loosely related conversations. In an agent environment, over-broad activation can route sensitive user inputs into a high-stakes compliance workflow unnecessarily, leading to incorrect regulatory framing, privacy exposure, or inappropriate advice being surfaced.
Content: No source excerpt is available for this finding.

Whitespace Padding

Medium

Category: Prompt Injection
Confidence: 70% confidence
Finding: Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.
Content: Scanner excerpt · SKILL.md (reported line 42)May include surrounding context.
md
## Step 2: Categorise | # | Category | Notification timeframe | | --- | ---------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------- | | 1 | Death of a participant | 24 hours | | 2 | Serious injury of a participant | 24 hours |

Whitespace Padding

Medium

Category: Prompt Injection
Confidence: 70% confidence
Finding: Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.
Content: Scanner excerpt · SKILL.md (reported line 44)May include surrounding context.
md
| # | Category | Notification timeframe | | --- | ---------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------- | | 1 | Death of a participant | 24 hours | | 2 | Serious injury of a participant | 24 hours | | 3 | Abuse or neglect of a participant (including alleged) | 24 hours | | 4 | Unlawful sexual or physical contact/assault of a participant | 24 hours |

Whitespace Padding

Medium

Category: Prompt Injection
Confidence: 70% confidence
Finding: Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.
Content: Scanner excerpt · SKILL.md (reported line 45)May include surrounding context.
md
| # | Category | Notification timeframe | | --- | ---------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------- | | 1 | Death of a participant | 24 hours | | 2 | Serious injury of a participant | 24 hours | | 3 | Abuse or neglect of a participant (including alleged) | 24 hours | | 4 | Unlawful sexual or physical contact/assault of a participant | 24 hours |

Whitespace Padding

Medium

Category: Prompt Injection
Confidence: 70% confidence
Finding: Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.
Content: Scanner excerpt · SKILL.md (reported line 46)May include surrounding context.
md
| # | Category | Notification timeframe | | --- | ---------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------- | | 1 | Death of a participant | 24 hours | | 2 | Serious injury of a participant | 24 hours | | 3 | Abuse or neglect of a participant (including alleged) | 24 hours | | 4 | Unlawful sexual or physical contact/assault of a participant | 24 hours |

Whitespace Padding

Medium

Category: Prompt Injection
Confidence: 70% confidence
Finding: Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.
Content: Scanner excerpt · SKILL.md (reported line 47)May include surrounding context.
md
| # | Category | Notification timeframe | | --- | ---------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------- | | 1 | Death of a participant | 24 hours | | 2 | Serious injury of a participant | 24 hours | | 3 | Abuse or neglect of a participant (including alleged) | 24 hours | | 4 | Unlawful sexual or physical contact/assault of a participant | 24 hours |

Whitespace Padding

Medium

Category: Prompt Injection
Confidence: 70% confidence
Finding: Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.
Content: Scanner excerpt · SKILL.md (reported line 48)May include surrounding context.
md
| # | Category | Notification timeframe | | --- | ---------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------- | | 1 | Death of a participant | 24 hours | | 2 | Serious injury of a participant | 24 hours | | 3 | Abuse or neglect of a participant (including alleged) | 24 hours | | 4 | Unlawful sexual or physical contact/assault of a participant | 24 hours |

Scope Creep

Low

Category: Excessive Agency
Confidence: 75% confidence
Finding: Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.
Content: Scanner excerpt · SKILL.md (reported line 122)May include surrounding context.
md
- **Never fabricate details not provided by the user** (no invented timestamps, witness names, or outcomes) to "complete" a report. Leave fields explicitly marked as `[TO BE CONFIRMED]` instead. - **Always flag the legal-advice boundary.** For genuinely ambiguous, high-severity, or legally contested cases (e.g. potential criminal conduct, disputed restrictive practice authorisation), explicitly recommend the user contact their compliance lead or the NDIS Commission directly rather than relying solely on the skill's categorisation. - **Never include full participant names, addresses, or other identifying details in draft outputs** unless the user has explicitly provided them for a real, final submission — default to initials/ID placeholders in working drafts. - **Do not extend scope into Behaviour Support Plan territory.** If an incident involves a registered restrictive practice under an authorised BSP, note that BSP-specific assessment is out of scope and refer to a registered Behaviour Support Practitioner. - **Do not let urgency override accuracy.** Even under the 24-hour deadline, never recommend submitting a notification with guessed or assumed facts — submit with whatever is genuinely known, clearly marked as preliminary. ---

Static analysis

No suspicious patterns detected.