tldr
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the git repository and conversation history to generate summaries, which could contain malicious instructions.
- Ingestion points: Reads conversation transcript and output from git status, git log, and git diff via SKILL.md and REFERENCE.md.
- Boundary markers: No specific delimiters or ignore embedded instructions warnings are defined for the input data processing, although the output follows a structured markdown format.
- Capability inventory: The skill uses git commands and jq for processing. It is explicitly restricted from writing files, committing changes, or running build/test tools in SKILL.md.
- Sanitization: The SKILL.md contains a specific guardrule to scan for and omit secrets, tokens, or environment dumps from the generated summary.
- [DYNAMIC_EXECUTION]: The skill executes a bundled configuration resolver script to manage settings.
- Evidence: Invokes templates/resolve-config.sh to process .tituskirch-skills.json using jq, as described in REFERENCE.md.
- [SAFE]: The documentation references an external JSON schema for configuration validation.
- Evidence: Links to a schema file hosted on the vendor's official GitHub repository at raw.githubusercontent.com/TitusKirch/skills/main/tituskirch-skills.schema.json.
Audit Metadata