How do I install the Joint Counter Autonomy Model Poisoning And Drift Cell skill?

Run npx skillmds add zwright8/joint-counter-autonomy-model-poisoning-and-drift-cell in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Which AI agents work with Joint Counter Autonomy Model Poisoning And Drift Cell?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Who published Joint Counter Autonomy Model Poisoning And Drift Cell?

zwright8 (@zwright8) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Joint Counter Autonomy Model Poisoning And Drift Cell by zwright8

Name: Joint Counter Autonomy Model Poisoning And Drift Cell
Rating: 4.2 (5 reviews)
Author: zwright8

Joint Counter Autonomy Model Poisoning And Drift Cell

Detect and mitigate poisoning, drift, and adversarial manipulation in deployed autonomy models across domains. Use when autonomous systems must remain mission-safe under active model-space attack.

zwright8 d8ea6ee 2 files · 2.6 KB Updated Sep 10, 2026

File contents

Joint Counter-Autonomy Model Poisoning and Drift Cell

Mission Scope

Treat this skill as planning and decision support for U.S. warfighter operations.
Confirm authority, data classification, response timeline, and engagement constraints.
Keep outputs unclassified unless handling guidance is provided.

Workflow

Establish mission objective, constraints, and commander decision horizon.
Fuse tool outputs into confidence-ranked options with explicit assumptions.
Build primary, alternate, and degraded courses of action with trigger conditions.
Bind each recommendation to authority gates, acknowledgments, and staff ownership.

Required Output Format

Situation snapshot.
Recommended primary action path.
Alternate/degraded path.
Decision points and authorities.
Staff tasking and suspense.

Domain Products

Primary products: mission posture summary, risk-to-mission matrix, execution branch plan.

External Tools and Protocol Integration

Use ../_shared/references/external-tools-protocols.md and ../_shared/references/tool-protocol-playbooks.md.
Use packet template DPL-AUTON-MODEL-INTEGRITY-001 from ../_shared/references/domain-tool-packet-library.md.
Bind tool and protocol choices to ts-autonomy-model-integrity-v1 from ../_shared/references/warfighter-external-tool-and-protocol-catalog.md.
Include transport/profile mapping and UTC freshness in all machine-to-machine exchanges.

Guardrails

Separate observed facts, assessed confidence, and unknowns.
Flag single-source claims and require corroboration for high-impact recommendations.
Keep human command approval checkpoints explicit for lethal or strategically escalatory actions.

Domain Toolchain Defaults

Primary: tool_suite_id=ts-autonomy-model-integrity-v1 with protocol_stack_id=ps-joint-tactical-link-stack-v1.
Alternate: tool_suite_id=ts-intel-fusion-v1 with protocol_stack_id=ps-cop-event-sharing-stack-v1.
Degraded: authenticated text or voice reporting with UTC acknowledgment chain.

Domain Packet Defaults

Default packet ID: DPL-AUTON-MODEL-INTEGRITY-001.
If packet scope mismatches mission constraints, define a provisional packet and assign validation owner/suspense.

---name: joint-counter-autonomy-model-poisoning-and-drift-celldescription: Detect and mitigate poisoning, drift, and adversarial manipulation in deployed autonomy models across domains. Use when autonomous systems must remain mission-safe under active model-space attack.---# Joint Counter-Autonomy Model Poisoning and Drift Cell## Mission Scope- Treat this skill as planning and decision support for U.S. warfighter operations.- Confirm authority, data classification, response timeline, and engagement constraints.- Keep outputs unclassified unless handling guidance is provided.## Workflow1. Establish mission objective, constraints, and commander decision horizon.2. Fuse tool outputs into confidence-ranked options with explicit assumptions.3. Build primary, alternate, and degraded courses of action with trigger conditions.4. Bind each recommendation to authority gates, acknowledgments, and staff ownership.## Required Output Format1. Situation snapshot.2. Recommended primary action path.3. Alternate/degraded path.4. Decision points and authorities.5. Staff tasking and suspense.## Domain ProductsPrimary products: mission posture summary, risk-to-mission matrix, execution branch plan.## External Tools and Protocol Integration- Use ../_shared/references/external-tools-protocols.md and ../_shared/references/tool-protocol-playbooks.md.- Use packet template DPL-AUTON-MODEL-INTEGRITY-001 from ../_shared/references/domain-tool-packet-library.md.- Bind tool and protocol choices to ts-autonomy-model-integrity-v1 from ../_shared/references/warfighter-external-tool-and-protocol-catalog.md.- Include transport/profile mapping and UTC freshness in all machine-to-machine exchanges.## Guardrails- Separate observed facts, assessed confidence, and unknowns.- Flag single-source claims and require corroboration for high-impact recommendations.- Keep human command approval checkpoints explicit for lethal or strategically escalatory actions.## Domain Toolchain Defaults- Primary: tool_suite_id=ts-autonomy-model-integrity-v1 with protocol_stack_id=ps-joint-tactical-link-stack-v1.- Alternate: tool_suite_id=ts-intel-fusion-v1 with protocol_stack_id=ps-cop-event-sharing-stack-v1.- Degraded: authenticated text or voice reporting with UTC acknowledgment chain.## Domain Packet Defaults- Default packet ID: DPL-AUTON-MODEL-INTEGRITY-001.- If packet scope mismatches mission constraints, define a provisional packet and assign validation owner/suspense.

Frequently asked questions