At a glance
What it does
A DSH web plugin for saving, finding, inserting, polishing, and optionally turning reusable prompts into Skills.
Web Profile
Not declared in supplied evidence
Evidence-verified
Checked Sep 10, 2026, 1:58 PM UTC
Code-evidenced contributions
What it adds to DSH
Adds a prompt-library panel plus # selection and AI Polish controls to the DSH web chat experience.
Mechanism evidence ↗Can generate linked DSH Skills from selected saved prompts, including inferred English names and descriptions.
Mechanism evidence ↗Before you choose it
Adds a Prompt Library button and # trigger to the DSH web chat bar. Save prompts with titles, bodies, and tags; search and reuse them by appending, replacing a draft, or filling variables and sending. It also offers AI Polish, optional automatic prompt learning, a user-maintained SOUL.md persona reference, and batch prompt-to-Skill generation.
Best for
DSH web users who repeatedly use structured prompts and want a local library for reuse and refinement.
Common tasks
- Keep tagged, searchable versions of recurring prompts.
- Insert a saved prompt into a draft, replace the draft, or fill template variables before sending.
- Use # in chat to filter and select a saved prompt quickly.
- Generate linked DSH Skills from selected prompts for later slash-command or model matching use.
Permissions and data
The supplied documentation describes local prompt-library storage and AI diagnostic logging; it does not document network destinations or credential requirements.
Permissions- Writes local SQLite prompt data and configuration beneath ~/.dsh/prompt-library/.
- Writes generated Skills beneath ~/.dsh/skills/<name>/.
- Stores prompts in SQLite at ~/.dsh/prompt-library/db/prompts.db.
- Stores daily AI diagnostic logs under ~/.dsh/prompt-library/log/.
- Stores the user-maintained SOUL.md persona file under ~/.dsh/prompt-library/character/.
- Regenerating a Skill for the same prompt overwrites that Skill directory.
- No external service endpoint is declared in the supplied evidence.
- No credential requirement is declared in the supplied evidence.
Limitations
- The pinned Git bundle is verified, but the specified npm package version was not found during supplied distribution checks.
- No Harness version range is declared in the supplied evidence.
- AI Polish, automatic learning, and generated-skill behavior are described by the README but were not executed in the supplied record.
- Requires Node.js >=22.19 according to the package manifest; peer dependencies include Cordis, React, and optionally dsh-llm/react-dom.
What DSHub checked
- The package is an MIT-licensed DSH bundle with a verified patch that inserts @sunjuntao/dsh-prompt-library as prompt-library.
- The package declares a web client injection and Node.js >=22.19.
- An immutable Git bundle path is verified at the supplied source commit.
What DSHub did not check
- Successful installation or operation in a DSH web profile was not demonstrated.
- Registry availability for @sunjuntao/dsh-prompt-library version 0.13.2 was not demonstrated.
- Runtime behavior, AI behavior, and compatibility with a particular Harness release were not demonstrated.
Pinned install
Install DSH Prompt Library
This plugin bundle does not have a DSH Plugin install action. Use its source documentation for the delivery method.
Maintainer source
Project README
🇨🇳 中文 | 🌐 English
</div>dsh-prompt-library
DSH (DeepSeek Harness) prompt library plugin: provides prompt management, AI polish and persona customization in the chat bar, helping you accumulate, reuse and continuously improve prompts.
The plugin treats conversation-level prompts as its core. Every feature is designed around this capability, while other auxiliary features are just a bonus. We warmly welcome users to share ideas and suggestions, and will keep iterating based on them.
Key Features
Prompt Library
- Manage frequently-used prompts (title + body + tags), with search, sorting, tag grouping and usage-count statistics
- The right panel shares the same style as the host's left sidebar, automatically squeezing and shrinking the chat area, freely toggling between expand/collapse
- Insert: appends to the existing content in the input box
- Overwrite: directly replaces the whole draft with this prompt
- Insert & Send: fills in template variables and sends with one click (uses the variable dialog when it contains
{{}}); outside#scenarios it is only available when the draft is empty, avoiding accidentally carrying existing content; when triggered by#, the trigger word is filtered out while the preceding text is kept and sent along - Type
#to quickly trigger selection with real-time filtering - The bottom shows the total number of tags and prompts in real time
Generate Skills
- Select prompts under "Prompt Library → Import/Export" to generate DSH Skills in batch
- AI generates an English skill name and description from the prompt content; the skill is written to
~/.dsh/skills/<name>/SKILL.md, and can be triggered by typing/skill-namein the chat box, or auto-matched by the model from the description - A link between the prompt and the skill is created automatically; regenerating the same prompt overwrites the original skill directory instead of adding endlessly
- If the body contains
{{variables}}, the "placeholder auto-fill" capability is marked at generation time — when using the skill, AI infers and fills them automatically from the current semantic context, no manual input needed
AI Polish
- The "AI Polish" button in the chat bar optimizes text with one click, then replaces it back into the input box with one click
- The polishing process follows the constraints of the persona file
Persona (AI read-only reference)
Each AI call and the whole chat session are constrained by a single SOUL.md persona file to form a stable assistant persona. The file is generated from a default template and rebuilt automatically if deleted; its content is manually maintained by the user, and AI only reads it without modifying it on its own.
| File | Meaning | Purpose |
|---|---|---|
SOUL.md |
Persona | Identity, tone/personality, working rules |
- Automatically learn input content into the prompt library, with AI smart-refining titles, tags, summaries and bodies
- The persona file is the user's explicit configuration (including the default template), and AI follows its personality, tone and working rules accordingly
Settings
Under DSH Settings → Prompt Library, adjust: auto learning, manual confirmation, AI smart polish, panel size, sidebar, button visibility, # triggering, etc. Changes take effect immediately.
Data Storage
The library uses SQLite (node:sqlite); all other configs and logs are stored under ~/.dsh/prompt-library/:
~/.dsh/prompt-library/
├── db/prompts.db # prompt library (SQLite)
├── log/
│ └── ai-YYYY-MM-DD.log # AI diagnostic logs (per-day files)
└── character/ # persona file
└── SOUL.md
Installation
dsh plugin --profile web add @sunjuntao/dsh-prompt-library
Usage
Start dsh web, click the "Prompt Library" button in the chat bar to open the panel, or type # to trigger it quickly; click "AI Polish" to polish your input with one click.
Development / Build
npm install
npm run deploy # type check + build + sync to DSH (restart dsh web to take effect)
Screenshots




Author
master1Sun
Operate deliberately
Install and manage
Prerequisites and target Profile
Target: Web Profile
Delivery: Dsh Bundle Git — master1Sun/dsh-prompt-library#1af79df60c871f6c61388b8e483d7b906b81aa7e。
Verify, update, and remove
Show lifecycle commands
dsh plugin --profile web listCompatibility and access
DSH web bundle; Node.js >=22.19 declared: Not declared in supplied evidence。
Review compatibility evidence ↗
Risk facts
Stores prompts, AI diagnostic logs, and persona configuration under ~/.dsh/prompt-library/.
Evidence ↗Regenerating a Skill for the same prompt overwrites its existing ~/.dsh/skills/<name>/ directory.
Evidence ↗Evidence and editorial reviewManifest, Bundle patch, distribution and freshness
Immutable evidence
Review status and source activity
Use the pinned Git source when reviewing this artifact; treat the README’s feature and storage descriptions as publisher claims, not runtime test results.
AI reviewed Sep 10, 2026, 1:59 PM UTC。GitHub facts last checked Sep 10, 2026, 1:59 PM UTC。
No material source change has been recorded since this evidence baseline.