Incomplete execution
Agent claims completion while subtasks are skipped, stubbed, or silently dropped.
Description
The agent reports success but has not actually done the work. Common forms: "TODO" comments left in generated code, function bodies replaced with `pass` or `...`, partial feature implementation while declaring "done", skipping error paths, or terminating before all user-specified items are addressed. Overconfidence paired with under-delivery. Distinct from scope_creep (which over-reaches beyond the request) — this is the opposite failure of under-delivering inside the requested surface.
Detection signals
- Grep for `TODO`, `FIXME`, `XXX`, `NotImplementedError`, or bare `pass`/`...` in the generated diff.
- Tests for features the user explicitly named don't exist.
- Agent's final message claims a subtask is done but the file contains only a stub.
- Function docstrings describe a behavior not implemented by the body.
Prior work
- Mehil B. Shah (2026) — Characterizing Faults in Agentic AI: A Taxonomy of Types, Symptoms, and Root Causes
- Melis Cemri (2025) — Why Do Multi-Agent LLM Systems Fail? MAST: A Multi-Agent System Failure Taxonomy
Crosswalks
Agent claims completion while key subtasks remain undone.
DAPLab "Incomplete Task Execution" is the direct counterpart.
Agent ends a task before completion criteria are met.
MAST "Premature Termination" is the direct counterpart at the multi-agent level.
Related failure modes
- Scope creep — Actions outside declared task boundary.
- Logic error — Correct API use, incorrect logic.
Mitigating tools
- Coverage.py — Code coverage measurement for Python. Detects test-manipulation by surfacing uncovered code.
- coze-loop — Full-lifecycle AI agent management platform with debugging, evaluation, and monitoring.
- lmnr — Laminar — open-source observability platform purpose-built for AI agents.
- opik — Debug, evaluate, and monitor LLM applications, RAG systems, and agentic workflows.
- pytest — Python test framework — discipline gate for "done means tests pass."
- voltagent — TypeScript-native AI agent engineering platform with built-in observability.