openspec-archive-change
Pass
Audited by Gen Agent Trust Hub on Nov 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands (
mkdir,mv) using identifiers provided by the user (the 'change name'). If the input is not properly sanitized by the agent platform, it could potentially lead to command injection if a malicious name containing shell metacharacters is used. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from local files which could contain malicious instructions.
- Ingestion points:
tasks.mdandopenspec/changes/<name>/GLOSSARY.md(specifically reading task completion status and glossary terms). - Boundary markers: Absent; the skill does not use specific delimiters or instructions to ignore embedded prompts in these files.
- Capability inventory: The skill has capabilities to execute shell commands (
mkdir,mv), call theopenspecCLI, and write to the project-levelGLOSSARY.mdfile. - Sanitization: Absent; the skill performs direct parsing and merging of terms without visible sanitization or validation of the content being merged.
Audit Metadata