Use ANOLISA when its terminal, context-efficiency, observability, or runtime features fit your agent stack; do not assume a DSH-specific installation or compatibility path from the README.
DSH integration
Ecosystem-related
Safety audit
Unaudited
Last verified
2026-09-13
License
Apache-2.0
01What can it help you accomplish?
Reduce redundant tool context for an existing AI-agent setup
Compressed tool schemas and responses before they reach the model, plus context reuse across sessions
Agent operators who want to control tool-context volume without replacing their shell, framework, or sandbox
Run agent work from a terminal while retaining familiar shell behavior
An AI-native Linux terminal with intent understanding, tool and Skill use, and approvals for risky work
Linux terminal users who want an agent interface without moving to a separate chat app
02How to install into DeepSeek Harness
Not specified by the author
03DSH integration and capability boundaries
Terminal agent entry with cosh-ng
Shell commands and natural-language requests in a Linux terminal→An agent-assisted terminal that can understand intent, use tools and Skills, and request approval for risky work
Token-less context compression
Tool schemas and tool responses headed to a model→Reduced redundancy before tool content reaches the model
The README says selected debug, trace, null, and empty values are dropped; dropped array items remain retrievable through a marker.Agent execution visibility
A Linux agent run→AgentSight traces covering user input, model and tool calls, token use, and sub-agent branches
Runtime boundaries and recovery points
Risky agent operations and workspace changes→Isolation through Agent Sec Core and workspace recovery points through ws-ckpt
04Who is it for? When not to use it?
Good for
- Agent operators who want to control tool-context volume without replacing their shell, framework, or sandbox
- Linux terminal users who want an agent interface without moving to a separate chat app
Not for
- The fetched README describes ANOLISA as an agent operating layer that can retain an existing shell, framework, and sandbox, but it does not document a DeepSeek Harness adapter, profile bundle, or activation command. Treat DSH integration as unverified.
- The README documents AgentSight's eBPF observation path specifically on Linux, so that visibility mechanism is not presented as a cross-platform feature.
- The README says cosh-ng is installed in system mode and shows a `sudo` installation example, so deploying that component can require elevated system privileges.
05Compatibility, maintenance and safety notes
- The fetched README describes ANOLISA as an agent operating layer that can retain an existing shell, framework, and sandbox, but it does not document a DeepSeek Harness adapter, profile bundle, or activation command. Treat DSH integration as unverified.
- The README documents AgentSight's eBPF observation path specifically on Linux, so that visibility mechanism is not presented as a cross-platform feature.
- Token-less savings concern tool responses entering model context rather than the total session bill; actual impact varies with the workload.
- The README says cosh-ng is installed in system mode and shows a `sudo` installation example, so deploying that component can require elevated system privileges.
Apache-2.0 source; author README statements captured from the fetched main branch (latest release sight/v0.12.1, 2026-09-10)
06Frequently asked questions
Is ANOLISA a native DeepSeek Harness integration?
No native DSH integration is documented in the fetched README. It is listed as ecosystem-level because the README describes a general agent operating layer, not a DSH adapter, profile bundle, or activation command.
Can I install ANOLISA into a DSH profile from this record?
No. The README includes ANOLISA installation examples, but none names DeepSeek Harness or documents DSH profile installation, verification, or rollback. Those DSH-specific fields are intentionally empty.
Will Token-less reduce a DSH session bill?
The README says Token-less removes redundancy before tool content reaches the model, but its stated savings apply to tool responses entering context rather than the whole session bill. No DSH-specific measurement is documented.
Does Token-less require agent-framework code changes?
The README says compression runs between the agent and model without agent-framework code changes. It does not specifically verify that behavior for DeepSeek Harness, so treat DSH compatibility as unverified.
Can DSH use ANOLISA's security and rollback components?
The README describes Agent Sec Core for isolating risky operations and ws-ckpt for workspace recovery points, but it does not document their connection to DeepSeek Harness. Validate integration independently before relying on them in DSH workflows.
07Related DSH workflows
dsh-routing-suite
by yjh051108
dsh-routing-suite — injector + router-standard kit: install the runtime injector first, then the task-aware reasoning-mode router preset (measured P1-P23).
brooks-lint
by hyhmrright
AI code reviews grounded in 12 classic engineering books — decay risk diagnostics with book citations, severity labels, and 6 analysis modes including full-sweep auto-fix
dsh-plugin-shop
by livxue
The most comprehensive DeepSeek Harness plugin market — refreshed daily, sourced across the Internet, reviewed before publishing.
dsh-hub-cli
by pax-beehive
Open-source CLI, schemas, resolver, and DSH agent tools for DSH Plugin Hub
08Data and sources
This page is generated from the project’s public documentation, repository metadata and a structured parse of DSH Plugins; last verified on 2026-09-13. Found an error? Submit a correction.
Best DeepSeek Harness Plugins
Twelve plugins worth installing first — picked from the whole catalog, across every category.