The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on the execution of several project-local shell scripts to manage problem reports, generate identifiers, and verify project traceability. These scripts include:\n
.guardrails/scripts/find-items.sh for listing and showing problem reports.\n
.guardrails/scripts/new-id.sh for minting unique identifiers.\n
check-trace.sh for validating the status and age of problem reports.\n
Workflow commands such as grill-requirements, analyze-risks, design-architecture, check-traceability, verify-before-merge, and merge-change.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes data from externally-controllable problem report files, which creates a potential surface for indirect prompt injection if those files contain malicious instructions.\n
Ingestion points: The agent reads markdown files from the docs/problems/ directory and configuration settings from .guardrails/config.yaml.\n
Boundary markers: The workflow enforces a structured format for problem report items (e.g., using status:, opened:, and disposition: keys) and uses indentation to delimit definitions, which helps separate data from instructions.\n
Capability inventory: The skill has the capability to write to the file system (recording problems), execute multiple shell scripts for project management, and trigger build/merge gates.\n
Sanitization: The skill incorporates validation via check-trace.sh, which checks for missing fields (INCOMPLETE-PROBLEM) and malformed dates (MALFORMED-DATE), providing some level of input verification.

resolve-problem