Back to skill

Security audit

BugDoctor: Diagnose the problem. Find the right fix.

Security checks for vulnerabilities and agentic risk

Overview

BugDoctor is a transparent debugging guide that stays scoped to user-authorized diagnosis and repair, with no hidden code, persistence, or credential-handling behavior found.

Install only if you want an agent to help diagnose software failures using local project evidence. Avoid pasting credentials or private payloads into public searches, and only authorize repairs when you are comfortable with the agent editing the affected project files.

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
Vulnerability Patterns
  • Memory PoisoningPersistent Context Injection, Context Window Stuffing, Memory Manipulation
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (8)

Persistent Context Injection

Medium
Category
Memory Poisoning
Confidence
80% confidence
Finding

Skill injects content designed to persist in agent memory or context across interactions. Persistent injection can alter agent behavior long after the initial interaction.

Content

Scanner excerpt · BUGDOCTOR.md (reported line 65)May include surrounding context.

md
For diagnosis, stop when the evidence supports a useful answer or a specific next check. For an authorized repair, continue until the failing behavior is resolved within scope or a concrete blocker requires missing evidence, access or a user decision. Reuse valid findings; revisit them when inputs change, a check fails or conflicting evidence appears. Another search, retry or audit needs a question whose answer could change the next action.

If asked to record an incident, use the existing format and include the trigger, target/version, decisive evidence, remedy, verification, relevant failed attempts and exclusions. Running this skill alone does not authorize persistent instruction changes.

If browsing or source access is unavailable, use supplied or cached evidence with its date and state the relevant access gap. A failed search does not establish that no issue exists. Limited evidence can still support a partial explanation or a useful check.

Persistent Context Injection

Medium
Category
Memory Poisoning
Confidence
80% confidence
Finding

Skill injects content designed to persist in agent memory or context across interactions. Persistent injection can alter agent behavior long after the initial interaction.

Content

Scanner excerpt · SKILL.md (reported line 70)May include surrounding context.

md
For diagnosis, stop when the evidence supports a useful answer or a specific next check. For an authorized repair, continue until the failing behavior is resolved within scope or a concrete blocker requires missing evidence, access or a user decision. Reuse valid findings; revisit them when inputs change, a check fails or conflicting evidence appears. Another search, retry or audit needs a question whose answer could change the next action.

If asked to record an incident, use the existing format and include the trigger, target/version, decisive evidence, remedy, verification, relevant failed attempts and exclusions. Running this skill alone does not authorize persistent instruction changes.

If browsing or source access is unavailable, use supplied or cached evidence with its date and state the relevant access gap. A failed search does not establish that no issue exists. Limited evidence can still support a partial explanation or a useful check.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The README advertises highly generic invocation phrases such as "BugDoctor this error," "Use BugDoctor on this," and support for a shortcode, which can overlap with ordinary troubleshooting language in unrelated conversations. In agent frameworks that auto-route on broad triggers, this can cause unintended activation of a powerful debugging/repair skill, expanding access to code, logs, runtime evidence, and possible repair actions without sufficiently explicit user intent.

Content

No source excerpt is available for this finding.

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Content

Scanner excerpt · REFERENCES.md (reported line 42)May include surrounding context.

md
- [Bug Diagnosis](https://clawhub.ai/mohitagw15856/skills/bug-diagnosis)
- [OpenClaw troubleshooting](https://clawhub.ai/playgod2023/skills/openclaw-troubleshooting)
- [OpenClaw Power Ops](https://clawhub.ai/kapslap/skills/openclaw-power-ops)
- [Superpowers systematic debugging](https://github.com/obra/superpowers/blob/main/skills/systematic-debugging/SKILL.md)
- [Chrome DevTools MCP troubleshooting](https://github.com/ChromeDevTools/chrome-devtools-mcp/blob/main/skills/troubleshooting/SKILL.md)

## Documentation

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Content

Scanner excerpt · REFERENCES.md (reported line 43)May include surrounding context.

md
- [OpenClaw troubleshooting](https://clawhub.ai/playgod2023/skills/openclaw-troubleshooting)
- [OpenClaw Power Ops](https://clawhub.ai/kapslap/skills/openclaw-power-ops)
- [Superpowers systematic debugging](https://github.com/obra/superpowers/blob/main/skills/systematic-debugging/SKILL.md)
- [Chrome DevTools MCP troubleshooting](https://github.com/ChromeDevTools/chrome-devtools-mcp/blob/main/skills/troubleshooting/SKILL.md)

## Documentation

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Content

Scanner excerpt · REFERENCES.md (reported line 80)May include surrounding context.

md
- [Evaluating AGENTS.md: Are Repository-Level Context Files Helpful for Coding Agents?](https://arxiv.org/abs/2602.11988v3)
- [Coding Agents Don't Know When to Act](https://arxiv.org/abs/2605.07769)
- [Superpowers condition-based waiting](https://github.com/obra/superpowers/blob/main/skills/systematic-debugging/condition-based-waiting.md)
- [Empirical Diagnostician](https://github.com/nivlewd1/prompt-optimizer/blob/main/skill/empirical-diagnostician/SKILL.md)
- [Debug Skill runtime capture](https://github.com/vltansky/debug-skill/blob/main/SKILL.md)
- [Chrome DevTools memory diagnosis](https://github.com/ChromeDevTools/chrome-devtools-mcp/blob/main/skills/memory-leak-debugging/SKILL.md)
- [AgentDebugX skill and trajectory diagnosis](https://github.com/AgentDebugX/AgentDebugX/blob/main/integrations/claude-code/plugins/agentdebug/skills/agentdebug/SKILL.md)

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Content

Scanner excerpt · REFERENCES.md (reported line 82)May include surrounding context.

md
- [Superpowers condition-based waiting](https://github.com/obra/superpowers/blob/main/skills/systematic-debugging/condition-based-waiting.md)
- [Empirical Diagnostician](https://github.com/nivlewd1/prompt-optimizer/blob/main/skill/empirical-diagnostician/SKILL.md)
- [Debug Skill runtime capture](https://github.com/vltansky/debug-skill/blob/main/SKILL.md)
- [Chrome DevTools memory diagnosis](https://github.com/ChromeDevTools/chrome-devtools-mcp/blob/main/skills/memory-leak-debugging/SKILL.md)
- [AgentDebugX skill and trajectory diagnosis](https://github.com/AgentDebugX/AgentDebugX/blob/main/integrations/claude-code/plugins/agentdebug/skills/agentdebug/SKILL.md)

These sources informed the design. Published tool or framework results are separate from BugDoctor's comparisons.

Skill Enumeration

Medium
Category
Agent Snooping
Confidence
80% confidence
Finding

Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.

Content

Scanner excerpt · REFERENCES.md (reported line 83)May include surrounding context.

md
- [Empirical Diagnostician](https://github.com/nivlewd1/prompt-optimizer/blob/main/skill/empirical-diagnostician/SKILL.md)
- [Debug Skill runtime capture](https://github.com/vltansky/debug-skill/blob/main/SKILL.md)
- [Chrome DevTools memory diagnosis](https://github.com/ChromeDevTools/chrome-devtools-mcp/blob/main/skills/memory-leak-debugging/SKILL.md)
- [AgentDebugX skill and trajectory diagnosis](https://github.com/AgentDebugX/AgentDebugX/blob/main/integrations/claude-code/plugins/agentdebug/skills/agentdebug/SKILL.md)

These sources informed the design. Published tool or framework results are separate from BugDoctor's comparisons.