---
title: "termination-self-monitoring"
description: "Checkable-reason discipline for stopping, deferring, narrowing or softening. Bundles the termination-self-monitoring skill (the 'what': a state-shaped reason, fatigue, clock, context budget, confidence as a feeling, diff"
canonical: https://agentpluginsdirectory.com/plugins/termination-self-monitoring
last-updated: 2026-09-16
---

# termination-self-monitoring
Checkable-reason discipline for stopping, deferring, narrowing or softening. Bundles the termination-self-monitoring skill (the 'what': a state-shaped reason, fatigue, clock, context budget, confidence as a feeling, difficulty as a mood, a run of apologies, is replaced by a checkable one: gate-not-run, owner-choice, budget-spent, limit-observed, or none and continue) with per-host hooks (the 'when') that scan the final message for those persona artifacts and for the [TERMINATION CHECK] block. Non-blocking by default; TERMMON_STRICT=1 blocks once. Portable core: the skill only; hooks are host-specific (hooks/ for Claude Code, cursor/ for Cursor).
- Slug: termination-self-monitoring
- Publisher: David Giordano
- Repository: https://github.com/3dgiordano/agent-plugins
- Manifest: plugins/termination-self-monitoring/plugin.json
- Version: 0.1.1
- License: Apache-2.0
- Category (editorial): other
- Skills: 1 (termination-self-monitoring)
- MCP servers: 0
- Stars: 0
- Repository created: 2026-09-15
- Repository last pushed: 2026-09-15
- Publisher type: User
- Listing: https://agentpluginsdirectory.com/plugins/termination-self-monitoring
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## What termination-self-monitoring does, in the publisher's words

Checkable-reason discipline for coding agents. When the agent stops, defers, narrows a task or softens a claim, the reason it gives must be one that can be checked, and a whole family of reasons it produces fluently cannot be: running out of context, this has been a long session, let's pick this up tomorrow, I'm not confident enough, given the complexity, a run of apologies. The plugin catches those phrases in the final message and asks for the substitution: which of four checkable reasons actually holds, or none, in which case the work continues.

It answers "is this stop justified by something checkable?": the mirror of persistence-self-monitoring, which answers "is it still worth insisting?". Persistence watches the agent stop too late, on no signal; this one watches it stop too early, on a signal it does not have.

From the project README, punctuation lightly normalized. Full text: https://raw.githubusercontent.com/3dgiordano/agent-plugins/HEAD/plugins/termination-self-monitoring/README.md

## Skills

- termination-self-monitoring: Checkable-reason discipline for stopping, deferring, narrowing or softening. A model inherits from its training data human-shaped reasons to stop, fatigue, a clock, a context budget, confidence as a feeling, difficulty as a mood, a run of apologies, that describe nothing it has. Each one is replace…

Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.