ClawHub
Back to skill

Security audit

PagerDuty On-Call

Security checks for vulnerabilities and agentic risk

Overview

This PagerDuty skill does what it says, but it can change live incident state with an API key and lacks clear safeguards or credential-handling guidance.

Install only if you are comfortable giving an agent PagerDuty credentials that can alter live incidents. Use the least-privileged PagerDuty token available, avoid storing it in shared shell profiles, rotate it regularly, and require explicit human confirmation before running acknowledge, resolve, trigger, snooze, or reassign commands.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (10)

Tainted flow: 'req' from os.environ.get (line 114, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Confidence
90% confidence
Finding

Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.

Content

Scanner excerpt · scripts/pd.py (reported line 40)May include surrounding context.

python
data = json.dumps(body).encode() if body else None
    req = urllib.request.Request(url, data=data, headers=headers or get_headers(), method=method)
    try:
        with urllib.request.urlopen(req) as resp:
            return json.loads(resp.read().decode())
    except urllib.error.HTTPError as e:
        error_body = e.read().decode()

Tainted flow: 'req' from os.environ.get (line 114, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Confidence
90% confidence
Finding

Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.

Content

Scanner excerpt · scripts/pd.py (reported line 121)May include surrounding context.

python
data = json.dumps(body).encode() if body else None
    req = urllib.request.Request(url, data=data, headers=headers or get_headers(), method=method)
    try:
        with urllib.request.urlopen(req) as resp:
            return json.loads(resp.read().decode())
    except urllib.error.HTTPError as e:
        error_body = e.read().decode()

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
94% confidence
Finding

The skill enables access to environment variables and networked PagerDuty operations but does not declare any explicit tool scope or permission boundaries. In an agent environment, this increases the chance of unintended credential access or remote actions being performed without clear policy controls.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The skill instructs users to export a long-lived PagerDuty API key and optional email identity without warning about secure storage, least privilege, or risk of disclosure. This can normalize unsafe handling of sensitive credentials in shell history, shared environments, or logs.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
78% confidence
Finding

The instructions guide the user to create a persistent API key without discussing lifecycle controls such as expiration, rotation, storage, or revocation. Long-lived credentials increase blast radius if exposed and are especially sensitive in agent-driven environments that can access environment variables.

Content

Scanner excerpt · SKILL.md (reported line 18)May include surrounding context.

export PAGERDUTY_FROM_EMAIL="oncall@yourcompany.com"

text

Get your API key: PagerDuty → User Icon → My Profile → User Settings → API Access Keys → Create New API Key

## Core Workflows

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill documents operationally destructive actions such as acknowledge, resolve, trigger, snooze, and reassign without warning that these change production incident state. An agent or user could execute them in the wrong context, causing missed alerts, false resolution, or disruption of on-call response.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 100)May include surrounding context.

bash
curl -s -H "Authorization: Token token=$PAGERDUTY_API_KEY" \
  -H "Accept: application/vnd.pagerduty+json;version=2" \
  "https://api.pagerduty.com/incidents?statuses[]=triggered&limit=25"

Key Concepts

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

These commands acknowledge or resolve live PagerDuty incidents immediately once invoked, with no confirmation step or explicit dry-run safeguard. In an agent-executed context, accidental invocation, prompt manipulation, or misunderstanding of user intent could silently alter incident state and disrupt operational response.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The trigger command can create new incidents in PagerDuty without any warning, confirmation, or contextual validation. In an autonomous or semi-autonomous agent setting, this increases the risk of false incident creation, alert fatigue, and operational disruption if the command is invoked due to ambiguous prompts or malicious instruction injection.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

Adding notes, snoozing incidents, and reassigning responders are all live operational changes, yet the tool performs them immediately with no confirmation or prominent warning. In the context of an agent skill specifically designed to operate PagerDuty, this raises the chance of unauthorized or accidental changes that can delay response, misroute ownership, or create misleading audit history.

Content

No source excerpt is available for this finding.