cursor-deeplinks
Generate Cursor deeplinks (prompt/command/rule and MCP install links) using the repo generator and the project output contract. Use when the user asks for a Cursor deeplink, cursor:// links, cursor.com/link links, or wants to create prompts/commands/rules via deeplinks.
What this skill does
# Cursor Deeplinks
This skill standardizes how to **create Cursor deeplinks** (per this repo’s conventions), including:
- prompt deeplinks
- command deeplinks (create `.cursor/commands/*.md` in the current project after approval)
- rule deeplinks (create `.cursor/rules/*.mdc` in the current project after approval)
- MCP install deeplinks
Primary reference in this repo: `doc/cursor-deeplinks.md`
Preferred generator: `scripts/cursor-deeplink-gen.py`
## Quick start (recommended)
Use the generator script to avoid encoding mistakes and to verify the **8,000 char URL limit** (post-encoding):
```bash
python3 scripts/cursor-deeplink-gen.py prompt --text "Create a README for this repo" --md
python3 scripts/cursor-deeplink-gen.py command --name "run-all-tests" --text "Run the full test suite and fix failures" --md
python3 scripts/cursor-deeplink-gen.py rule --name "python-style" --text "Prefer ruff/black; avoid implicit Any." --md
python3 scripts/cursor-deeplink-gen.py mcp-install --name postgres --config-file mcp.json --md
```
## Output contract (always follow)
When asked to produce a deeplink, return:
1. **WEB link (primary)**: `https://cursor.com/link/...`
2. **APP link (secondary)**: `cursor://anysphere.cursor-deeplink/...`
3. **Expected result**: what Cursor will create/show after the user approves (e.g. where the file will be created)
4. A short safety note: **“Check secrets + 8,000 URL limit”**
## Deeplink formats (what to generate)
Cursor supports two equivalent bases:
- **WEB**: `https://cursor.com/link`
- **APP**: `cursor://anysphere.cursor-deeplink`
Supported paths and params:
### Prompt
- Path: `/prompt`
- Params: `text=<promptText>`
### Command
- Path: `/command`
- Params: `name=<commandName>`, `text=<commandContent>`
- Expected result: Cursor will propose creating a command file under `.cursor/commands/` in the current project after approval.
### Rule
- Path: `/rule`
- Params: `name=<ruleName>`, `text=<ruleContent>`
- Expected result: Cursor will propose creating a rule file under `.cursor/rules/` in the current project after approval.
### MCP install
- Path: `/mcp/install`
- Params: `name=<serverName>`, `config=<base64(JSON.stringify(transportConfig))>`
- The `transportConfig` is the per-server object (e.g. `{"command":"npx","args":[...]}`), not the whole `mcp.json` mapping.
## Manual construction (only if you cannot run the generator)
1. Choose base (**WEB** preferred) + path.
2. Build query params.
3. **URL-encode** all params (including newlines/spaces).
4. Ensure total URL length (after encoding) \(\le 8000\).
Notes:
- URL encoding is fragile; prefer the generator.
- For MCP install, `config` must be base64 of the stringified JSON. Base64 output must be safe for query params (must be URL-encoded; `+` becomes `%2B`).
## Safety + UX checks (must do)
- **Secrets**: never embed API keys, tokens, passwords, private keys, or other sensitive content in deeplink text/config.
- **Length**: enforce the **8,000 character** max (post-encoding). If over limit, shorten the content or generate artifacts via normal repo edits instead of a deeplink.
- **Expectation**: deeplinks **do not auto-execute**; the user must review and approve in Cursor.
## Examples (copy/paste patterns)
### Prompt deeplink (generator)
```bash
python3 scripts/cursor-deeplink-gen.py prompt --text "Write a short changelog entry for the last commit" --md
```
### Command deeplink (generator)
```bash
python3 scripts/cursor-deeplink-gen.py command --name "audit-config" --text "Review mcp.json for risky settings and propose fixes." --md
```
### Rule deeplink (generator)
```bash
python3 scripts/cursor-deeplink-gen.py rule --name "cursor-deeplinks" --text "When asked for a deeplink, output WEB+APP links + expected result + safety note." --md
```
### MCP install deeplink (generator)
```bash
python3 scripts/cursor-deeplink-gen.py mcp-install --name "github" --config-file mcp.json --md
```
## Additional resources
- Repo docs: `doc/cursor-deeplinks.md`
- Official docs:
- `https://cursor.com/docs/integrations/deeplinks.md`
- `https://cursor.com/docs/context/mcp/install-links.md`
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.