accelint-qrspi-apply
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests potentially untrusted data from local files to influence the behavior of sub-agents. Ingestion points: reads parallelization strategies from tasks.md and project context from openspec/config.yaml. Capability inventory: spawns sub-agents and executes shell commands (git, python3, openspec). Boundary markers: employs <project_context> tags and XML comments to delimit injected context. Sanitization: relies on instructional guardrails to treat context as behavioral constraints rather than code content.
- [DYNAMIC_EXECUTION]: Steps 7 and 34 utilize python3 -c to execute a hardcoded string for generating ISO 8601 timestamps in design.md frontmatter.
- [COMMAND_EXECUTION]: The skill executes multiple local CLI tools including git, openspec, cat, and python3 for environment verification, configuration management, and implementation telemetry.
Audit Metadata