retrospective
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_CONTEXT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from multiple sources including sprint status YAML files, markdown plans, and project source code (via grep for TODO/FIXME comments). There are no explicit boundary markers or sanitization routines defined to handle these external inputs. As the skill uses this data to generate reports and potentially trigger follow-up actions like sprint planning, it is susceptible to instructions embedded within the processed project files.
- Ingestion points: production/sprint-status.yaml, production/sprints/, production/milestones/, and general codebase.
- Boundary markers: Absent.
- Capability inventory: Read, Glob, Grep, Write, Bash, AskUserQuestion.
- Sanitization: Absent.
- [DYNAMIC_CONTEXT_INJECTION]: The skill uses the dynamic context injection pattern in SKILL.md to execute a local configuration script (yaml-helper.sh) at load time. This command resolves environment-specific keys such as automation levels and workflow settings.
- [COMMAND_EXECUTION]: The skill executes git commands (git log) via the Bash tool to retrieve commit history for velocity and completion analysis. This is a standard operation for the skill's stated purpose.
Audit Metadata