apc-implement

Pass

Audited by Gen Agent Trust Hub on Nov 19, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The skill uses a relative path (../../.agents/workflows/apc-implement.md) to load instructions at runtime. This redirection mechanism moves the actual execution logic to a file located outside the skill's own directory.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests content from a local workspace file and treats it as authoritative instructions without sanitization.
  • Ingestion points: Reads content from .agents/workflows/apc-implement.md relative to the workspace root (SKILL.md).
  • Boundary markers: Absent; the agent is explicitly told to follow the file "exactly as written."
  • Capability inventory: While not defined in this file, the "PHASE 4 implementation" description implies the agent will perform file modifications or command executions based on the loaded instructions.
  • Sanitization: Absent; there is no validation of the target file's content before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 19, 2026, 01:17 AM