Security audit

Linear GraphQL Agent

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Linear GraphQL helper that can read or change Linear workspace data when configured with credentials, but the access is purpose-aligned and visible.

Install this only for agents you want to connect to a Linear workspace. Prefer read-only or least-privilege credentials when possible, avoid pasting literal API keys unless necessary, and confirm the target workspace and issue IDs before allowing create, update, archive, or comment mutations.

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
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (10)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 98% confidence
Finding: The declared description presents an operational skill for managing Linear resources through the Linear GraphQL API. However, the supplied code chunk is purely a local validation script for repository files and documentation standards. Its behavior is centered on checking file existence and required strings in markdown/YAML, not on executing Linear operations. This is a material mismatch in primary purpose and capabilities.

Credential Access

High

Category: Privilege Escalation
Content: --priority 100 ``` **Important (Personal API Key only):** Linear API keys require `Authorization: lin_api_XXX` format (no `Bearer ` prefix). The `--header "Authorization:{{secret}}"` configuration above is required for API key auth. For OAuth access tokens, use standard `Authorization: Bearer <token>` semantics (handled by the OAuth credential flow in `uxc`). Credential/binding IDs in this skill use `linear-graphql` by convention; IDs are local aliases and can be changed if needed.
Confidence: 70% confidence
Finding: Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Missing User Warnings

Medium

Confidence: 92% confidence
Finding: The documentation includes state-changing and destructive Linear mutations such as issue creation, update, archiving, and comment creation without any warning, confirmation guidance, or clear indication that these commands modify remote production data. In an agent skill context, examples are often copied verbatim or automated, so omission of safety guardrails increases the risk of unintended changes to real workspace data.

External Transmission

Medium

Category: Data Exfiltration
Content: rg -q '^description:\s*.+' "${SKILL_FILE}" || fail 'missing description' rg -q 'command -v linear-graphql-cli' "${SKILL_FILE}" || fail 'missing link-first command check' rg -q 'uxc link linear-graphql-cli https://api.linear.app/graphql' "${SKILL_FILE}" || fail 'missing fixed link create command' rg -q 'linear-graphql-cli -h' "${SKILL_FILE}" || fail 'missing host help example' rg -q 'linear-graphql-cli query/issues -h' "${SKILL_FILE}" || fail 'missing operation-level help example'
Confidence: 60% confidence
Finding: Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium

Category: Data Exfiltration
Content: rg -q '^description:\s*.+' "${SKILL_FILE}" || fail 'missing description' rg -q 'command -v linear-graphql-cli' "${SKILL_FILE}" || fail 'missing link-first command check' rg -q 'uxc link linear-graphql-cli https://api.linear.app/graphql' "${SKILL_FILE}" || fail 'missing fixed link create command' rg -q 'linear-graphql-cli -h' "${SKILL_FILE}" || fail 'missing host help example' rg -q 'linear-graphql-cli query/issues -h' "${SKILL_FILE}" || fail 'missing operation-level help example'
Confidence: 60% confidence
Finding: Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium

Category: Data Exfiltration
Content: rg -q '^description:\s*.+' "${SKILL_FILE}" || fail 'missing description' rg -q 'command -v linear-graphql-cli' "${SKILL_FILE}" || fail 'missing link-first command check' rg -q 'uxc link linear-graphql-cli https://api.linear.app/graphql' "${SKILL_FILE}" || fail 'missing fixed link create command' rg -q 'linear-graphql-cli -h' "${SKILL_FILE}" || fail 'missing host help example' rg -q 'linear-graphql-cli query/issues -h' "${SKILL_FILE}" || fail 'missing operation-level help example'
Confidence: 60% confidence
Finding: Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium

Category: Data Exfiltration
Content: rg -q '^description:\s*.+' "${SKILL_FILE}" || fail 'missing description' rg -q 'command -v linear-graphql-cli' "${SKILL_FILE}" || fail 'missing link-first command check' rg -q 'uxc link linear-graphql-cli https://api.linear.app/graphql' "${SKILL_FILE}" || fail 'missing fixed link create command' rg -q 'linear-graphql-cli -h' "${SKILL_FILE}" || fail 'missing host help example' rg -q 'linear-graphql-cli query/issues -h' "${SKILL_FILE}" || fail 'missing operation-level help example'
Confidence: 60% confidence
Finding: Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium

Category: Data Exfiltration
Content: rg -q '^description:\s*.+' "${SKILL_FILE}" || fail 'missing description' rg -q 'command -v linear-graphql-cli' "${SKILL_FILE}" || fail 'missing link-first command check' rg -q 'uxc link linear-graphql-cli https://api.linear.app/graphql' "${SKILL_FILE}" || fail 'missing fixed link create command' rg -q 'linear-graphql-cli -h' "${SKILL_FILE}" || fail 'missing host help example' rg -q 'linear-graphql-cli query/issues -h' "${SKILL_FILE}" || fail 'missing operation-level help example'
Confidence: 60% confidence
Finding: Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium

Category: Data Exfiltration
Content: rg -q '^description:\s*.+' "${SKILL_FILE}" || fail 'missing description' rg -q 'command -v linear-graphql-cli' "${SKILL_FILE}" || fail 'missing link-first command check' rg -q 'uxc link linear-graphql-cli https://api.linear.app/graphql' "${SKILL_FILE}" || fail 'missing fixed link create command' rg -q 'linear-graphql-cli -h' "${SKILL_FILE}" || fail 'missing host help example' rg -q 'linear-graphql-cli query/issues -h' "${SKILL_FILE}" || fail 'missing operation-level help example'
Confidence: 60% confidence
Finding: Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

External Transmission

Medium

Category: Data Exfiltration
Content: rg -q '^description:\s*.+' "${SKILL_FILE}" || fail 'missing description' rg -q 'command -v linear-graphql-cli' "${SKILL_FILE}" || fail 'missing link-first command check' rg -q 'uxc link linear-graphql-cli https://api.linear.app/graphql' "${SKILL_FILE}" || fail 'missing fixed link create command' rg -q 'linear-graphql-cli -h' "${SKILL_FILE}" || fail 'missing host help example' rg -q 'linear-graphql-cli query/issues -h' "${SKILL_FILE}" || fail 'missing operation-level help example'
Confidence: 60% confidence
Finding: Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Static analysis

No suspicious patterns detected.