setup-promise-to-proof
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill reads external, potentially untrusted data from the repository, including
AGENTS.md,CLAUDE.md,CONTEXT.md, anddocs/agents/, as well as GitHub label names via theghCLI. This data is used to draft configuration files and propose mappings. - Ingestion points:
AGENTS.md,CLAUDE.md,docs/agents/,CONTEXT.md,CONTEXT-MAP.md, GitHub repository labels. - Boundary markers: Not explicitly defined in the instructions for reading these files.
- Capability inventory: The skill writes to the local filesystem (
.p2p/,docs/agents/,.gitignore) and executes a Python script. - Sanitization: No explicit sanitization or validation of the content read from these files is mentioned.
- [COMMAND_EXECUTION]: The skill executes local commands and scripts to perform its setup tasks.
- Local script execution: It runs
python3 <skill-dir>/scripts/p2p_filesystem.py --repo <root> setupto manage directory structures and detect conflicts. - Tool usage: It uses the
gh(GitHub CLI) to inspect repository labels and recommendsgh label createcommands to the user. - Git operations: It uses
git check-ignore -vto debug ignore patterns.
Audit Metadata