Context Discipline

Framework stopping AI agents from taking dangerous actions based on missing information.

Install
cmdop skills install agensi-context-discipline

Context Discipline is a procedural skill for AI agents that enforces a strict, artifact-gated workflow at every step of execution. Rather than letting an agent silently skip steps, make unverified assumptions, or declare a task complete without evidence, this skill requires agents to inventory available context, declare assumptions explicitly, and produce concrete artifacts as proof before any workflow transition is allowed.

The core mechanism treats each step boundary as a gated edge: an agent cannot advance until it can supply the required evidence. If that evidence is missing — because context is incomplete or an operation has not been verified — the agent stops and requests clarification rather than guessing. This is a fail-closed design: ambiguity causes a halt, not a guess.

For destructive operations such as deletions or security changes, the skill mandates a human-in-the-loop approval gate before proceeding. For multi-step coding workflows, it requires that tests pass and changes are recorded before the agent claims completion. For agent handoffs, it standardizes a “state of the world” summary so the receiving agent has full, auditable context. Unverified actions are automatically labeled as partial or not_run rather than succeeded.

This skill is aimed at developers building complex agentic pipelines where auditability and safety are non-negotiable requirements. It is a structured decision matrix and artifact schema, not a simple prompt — making it more robust than ad-hoc instructions for controlling agent behavior in production workflows.

Use cases

  • Enforce verification gates in multi-step coding agents before marking tasks complete
  • Add mandatory human approval checkpoints before an agent executes destructive database or file operations
  • Standardize context handoff state between chained agents in a multi-agent pipeline
  • Automatically flag unverified or partially completed actions instead of reporting false success
  • Audit agent decision trails by requiring artifact evidence at every workflow transition
  • Prevent agents from proceeding on vague or incomplete instructions by forcing explicit assumption declarations

When to use it

  • Building agentic loops where silent assumptions or false completion claims are unacceptable
  • Any workflow that includes irreversible or destructive operations requiring human sign-off
  • Multi-agent pipelines where one agent's output becomes the next agent's input and context fidelity matters
  • Production environments where auditability of every agent action is required

When not to use it

  • Simple single-step tasks where workflow gating adds overhead with no safety benefit
  • Agents that do not require human-in-the-loop checkpoints and need fully autonomous uninterrupted execution
  • Use cases where no artifact or evidence schema can be defined for success verification