Skip to content
Back to search

parallel-agent-skills

SkillGitHubIncomplete

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

It has changed since its scan on 11 Nov 2026.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/parallel-web/parallel-agent-skills with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/parallel-web/parallel-agent-skills'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/parallel-web--parallel-agent-skills/badge.svg)](/listings/parallel-web--parallel-agent-skills)

HTML

<a href="/listings/parallel-web--parallel-agent-skills"><img src="/api/listings/parallel-web--parallel-agent-skills/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

Agent Skills

Use `parallel-cli` to install skills into Cursor, Cline, GitHub Copilot, and other compatible agents.

# Install all Parallel skills
parallel-cli skills install

Claude Code

Available as a [Claude Code Plugin Marketplace](https://code.claude.com/docs/en/discover-plugins).

/plugin marketplace add parallel-web/parallel-agent-skills
/plugin install parallel
# restart Claude Code before continuing!

After restarting, run `/mcp` to inspect the plugin-provided `parallel-search` server. Search, Fetch, and Decisions support anonymous use. Research requires Parallel sign-in; in clients that support tool-level authentication, requesting Research prompts you to connect your account. Once signed in, all authenticated MCP calls are billed at [Parallel API rates](https://parallel.ai/pricing).

Some Claude clients do not complete tool-level sign-in or expose Research before authentication. If that happens, use the connection's sign-in control. If the server supplies an authenticated fallback URL, review the account and billing implications before switching to it. Do not silently change endpoints or add a duplicate server. See the setup skill for the explicit-login fallback.

Upgrading from v0.10.0 changes the Claude plugin endpoint to unified MCP at `https://mcp.parallel.ai/mcp/claude`. Update the plugin, fully exit Claude Code, and start a fresh session. Existing conversations can retain old tool arguments after `/reload-plugins`; if a call rejects `model_name` or `session_id`, start a fresh session and use the current schema. An old endpoint's login may not carry over. If you previously added a separate Parallel connection, verify the new one works before disabling the duplicate.

For the CLI-backed skills (deep research, enrichment, FindAll, Monitor), install and authenticate the CLI:

# this will install/update CLI and authenticate if not done already
/parallel:parallel-cli-setup

OpenAI Codex

Install skills using the built-in skill installer (run inside Codex):

$skill-installer parallel-web/parallel-agent-skills

Then run the setup skill to install/auth the CLI:

/parallel:parallel-cli-setup

Google Antigravity

Install the standalone Search MCP plugin with the Antigravity CLI:

git clone https://github.com/parallel-web/parallel-agent-skills.git
cd parallel-agent-skills
agy plugin install ./plugins/antigravity/parallel

This package provides free web search and page fetching without a Parallel account, API key, or local Parallel CLI. Use the package directory above so Antigravity loads only the Search MCP connection. See the [Antigravity package README](plugins/antigravity/parallel/README.md) for project-local installation, validation, example prompts, and removal.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

.claude-plugin/marketplace.json · Manifest · 50 lines
  1. {
  2. "name": "parallel-agent-skills",
  3. "owner": {
  4. "name": "Parallel Web Systems",
  5. "email": "support@parallel.ai",
  6. "url": "https://parallel.ai"
  7. },
  8. "metadata": {
  9. "description": "Search the web, read pages, research questions, and assess claims through Parallel's MCP server. Search, Fetch, and Decisions work without an account; Research requires sign-in. Includes CLI skills for deep research, enrichment, FindAll, and monitoring.",
  10. "version": "0.12.0",
  11. "pluginRoot": "./"
  12. },
  13. "plugins": [
  14. {
  15. "name": "parallel",
  16. "description": "Search the web, read pages, research questions, and assess claims through Parallel's MCP server. Search, Fetch, and Decisions work without an account; Research requires sign-in. Includes CLI skills for deep research, enrichment, FindAll, and monitoring.",
  17. "version": "0.12.0",
  18. "source": "./",
  19. "author": {
  20. "name": "Parallel Web Systems",
  21. "email": "support@parallel.ai"
  22. },
  23. "homepage": "https://docs.parallel.ai/home",
  24. "repository": "https://github.com/parallel-web/parallel-agent-skills",
  25. "license": "MIT",
  26. "keywords": [
  27. "web-search",
  28. "research",
  29. "data-enrichment",
  30. "entity-discovery",
  31. "web-monitoring",
  32. "parallel",
  33. "parallel-cli",
  34. "ai-tools",
  35. "content-extraction"
  36. ],
  37. "category": "productivity",
  38. "tags": [
  39. "search",
  40. "web",
  41. "mcp",
  42. "research",
  43. "enrichment",
  44. "findall",
  45. "monitor",
  46. "cli"
  47. ]
  48. }
  49. ]
  50. }
README.md · README · 257 lines
  1. # Parallel Agent Skills
  2. Web search and page fetching for AI coding agents, through [Parallel](https://parallel.ai) — plus [Agent Skills](https://agentskills.io/specification) for deep research, data enrichment, entity discovery, and web monitoring.
  3. The plugin ships two things:
  4. - **A hosted MCP server** (`parallel-search`). In Claude, it provides anonymous `web_search`, `web_fetch`, and `web_decisions`, plus `web_research` with Parallel sign-in. Authenticated calls are billed to your Parallel account. The Codex and Antigravity packages retain anonymous, rate-limited search and fetch.
  5. - **CLI-backed skills** for deep research, enrichment, FindAll, and Monitor. These run through `parallel-cli` and need it installed, authenticated, and funded.
  6. You can use the first without ever setting up the second.
  7. ## Prerequisites
  8. Claude Search, Fetch, and Decisions work without a Parallel account, subject to anonymous usage limits. Research requires a Parallel account with sufficient balance. No API key or local CLI is needed for MCP tools. Codex and Antigravity retain anonymous search and fetch with no account required.
  9. The execution skills require `parallel-cli` (installed, authenticated, and funded). The [`parallel-cli-setup`](skills/parallel-cli-setup/SKILL.md) skill walks an agent through install, auth, balance, and skills install end-to-end — install the plugin/skills below, then run `/parallel:parallel-cli-setup` from your agent.
  10. `migrate-to-parallel` updates an application's own web-data integration. It uses the appropriate Parallel API or SDK and needs `PARALLEL_API_KEY` only for an explicitly authorized live smoke test.
  11. ## Installation
  12. ### Agent Skills
  13. Use `parallel-cli` to install skills into Cursor, Cline, GitHub Copilot, and other compatible agents.
  14. ```bash
  15. # Install all Parallel skills
  16. parallel-cli skills install
  17. ```
  18. ### Claude Code
  19. Available as a [Claude Code Plugin Marketplace](https://code.claude.com/docs/en/discover-plugins).
  20. ```bash
  21. /plugin marketplace add parallel-web/parallel-agent-skills
  22. /plugin install parallel
  23. # restart Claude Code before continuing!
  24. ```
  25. After restarting, run `/mcp` to inspect the plugin-provided `parallel-search`
  26. server. Search, Fetch, and Decisions support anonymous use. Research requires
  27. Parallel sign-in; in clients that support tool-level authentication, requesting
  28. Research prompts you to connect your account. Once signed in, all authenticated
  29. MCP calls are billed at [Parallel API rates](https://parallel.ai/pricing).
  30. Some Claude clients do not complete tool-level sign-in or expose Research
  31. before authentication. If that happens, use the connection's sign-in control.
  32. If the server supplies an authenticated fallback URL, review the account and
  33. billing implications before switching to it. Do not silently change endpoints
  34. or add a duplicate server. See the setup skill for the explicit-login fallback.
  35. Upgrading from v0.10.0 changes the Claude plugin endpoint to unified MCP at
  36. `https://mcp.parallel.ai/mcp/claude`. Update the plugin, fully exit Claude Code,
  37. and start a fresh session. Existing conversations can retain old tool arguments
  38. after `/reload-plugins`; if a call rejects `model_name` or `session_id`, start a
  39. fresh session and use the current schema. An old endpoint's login may not carry
  40. over. If you previously added a separate Parallel connection, verify the new
  41. one works before disabling the duplicate.
  42. For the CLI-backed skills (deep research, enrichment, FindAll, Monitor), install
  43. and authenticate the CLI:
  44. ```bash
  45. # this will install/update CLI and authenticate if not done already
  46. /parallel:parallel-cli-setup
  47. ```
  48. ### OpenAI Codex
  49. Install skills using the built-in skill installer (run inside Codex):
  50. ```text
  51. $skill-installer parallel-web/parallel-agent-skills
  52. ```
  53. Then run the setup skill to install/auth the CLI:
  54. ```text
  55. /parallel:parallel-cli-setup
  56. ```
  57. ### Google Antigravity
  58. Install the standalone Search MCP plugin with the Antigravity CLI:
  59. ```bash
  60. git clone https://github.com/parallel-web/parallel-agent-skills.git
  61. cd parallel-agent-skills
  62. agy plugin install ./plugins/antigravity/parallel
  63. ```
  64. This package provides free web search and page fetching without a Parallel account,
  65. API key, or local Parallel CLI. Use the package directory above so Antigravity
  66. loads only the Search MCP connection. See the [Antigravity package README](plugins/antigravity/parallel/README.md)
  67. for project-local installation, validation, example prompts, and removal.
  68. ## CDN and discovery
  69. A human + machine-readable catalog is published at [skills.parallel.ai](https://skills.parallel.ai).
  70. Useful endpoints:
  71. - [skills.parallel.ai](https://skills.parallel.ai) — human-friendly catalog and install instructions
  72. - [skills.parallel.ai/index.json](https://skills.parallel.ai/index.json) — machine-readable skill index
  73. - `https://skills.parallel.ai/<skill>/SKILL.md` — live raw skill file
  74. - `https://skills.parallel.ai/<skill>/manifest.json` — file manifest + checksums
  75. - `https://skills.parallel.ai/<skill>/versions.json` — release history for that skill
  76. - `https://skills.parallel.ai/archives/<skill>/<version>.zip` — immutable GitHub Release archive via CDN redirect
  77. ## Skills
  78. Skills follow the [Agent Skills](https://agentskills.io/specification) specification and double as Claude Code slash commands.
  79. | Skill | Description |
  80. | ---------------------------- | --------------------------------------------------------- |
  81. | **parallel-search-setup** | Verify and troubleshoot Parallel MCP; sign in for Research |
  82. | **parallel-mcp-setup** | Set up authenticated Parallel MCP connections; Bifrost Search MCP is the first supported path |
  83. | **parallel-web-search** | CLI-backed search for explicit CLI use or when MCP is unavailable |
  84. | **parallel-web-extract** | CLI-backed extraction for explicit CLI use or when MCP is unavailable |
  85. | **parallel-api** | Build on the Parallel API: choose an API, set up the key, write the integration, fix errors |
  86. | **choose-your-parallel-api** | Choose the right Parallel API and configuration |
  87. | **parallel-search-evaluation** | Benchmark web search providers on your own workload. |
  88. | **parallel-deep-research** | Comprehensive research and analysis |
  89. | **parallel-data-enrichment** | Enrich lists of companies, people, products |
  90. | **parallel-findall** | Discover entities matching a natural-language description |
  91. | **parallel-monitor** | Continuously track the web for changes (with webhooks) |
  92. | **parallel-memory** | Recall and manage saved Parallel runs |
  93. | **migrate-to-parallel** | Migrate Exa, Tavily, Perplexity, or Firecrawl integrations to Parallel |
  94. | **parallel-cli-setup** | Install/update CLI, authenticate, and handle balance |
  95. | **status** | Check running research task status |
  96. | **result** | Get completed research task result |
  97. ## Examples
  98. ```text
  99. /parallel:parallel-web-search latest React 19 features
  100. /parallel:parallel-search-evaluation compare search providers on production questions
  101. /parallel:parallel-web-extract https://docs.parallel.ai
  102. /parallel:parallel-deep-research competitive landscape of AI code assistants
  103. /parallel:parallel-data-enrichment Apple, Microsoft, Google - get CEO names
  104. /parallel:parallel-findall AI startups that raised Series A in 2026
  105. /parallel:parallel-monitor track price changes for the iPhone 16 Pro
  106. /parallel:parallel-memory retrieve past research about AI code assistants
  107. /parallel:parallel-api add Parallel web search to this app
  108. /parallel:migrate-to-parallel migrate this app from Tavily to Parallel
  109. /parallel:migrate-to-parallel migrate this app from Perplexity to Parallel
  110. /parallel:migrate-to-parallel migrate this app from Firecrawl to Parallel
  111. /parallel:parallel-cli-setup
  112. ```
  113. ## Accounts, billing, and free access
  114. ### Claude
  115. The Claude plugin connects to
  116. `https://mcp.parallel.ai/mcp/claude`.
  117. Search, Fetch, and Decisions accept anonymous calls. Research requires OAuth
  118. and usable account credits. After sign-in, authenticated calls are billed under
  119. your account's [pricing and limits](https://parallel.ai/pricing). Manage balance
  120. and usage at [platform.parallel.ai](https://platform.parallel.ai).
  121. MCP research returns a cited answer. The existing deep-research skill continues
  122. to run asynchronous Task API reports through the CLI; it is a separate workflow.
  123. ### Codex and other anonymous Search MCP installations
  124. The bundled `parallel-search` MCP server is anonymous by design: `web_search` and
  125. `web_fetch` work with no account, API key, or login, subject to free-tier rate
  126. limits. The free tier is intended for personal agents, hobby projects, and
  127. exploration.
  128. For higher search limits, usage analytics, and production workloads, create an
  129. account at [platform.parallel.ai](https://platform.parallel.ai), then add a
  130. separate authenticated Search MCP connection. The same service supports a
  131. Parallel API key on `https://search.parallel.ai/mcp` or OAuth through
  132. `https://search.parallel.ai/mcp-oauth`. Authenticated usage is billed under your
  133. account's pricing and limits — see
  134. [parallel.ai/pricing](https://parallel.ai/pricing).
  135. Keep the plugin-provided connection anonymous rather than editing the installed
  136. plugin. After the authenticated connection works, use `/mcp` to ensure only one
  137. Parallel Search connection is enabled; if both appear, toggle the bundled
  138. anonymous server off. See
  139. [`parallel-search-setup`](skills/parallel-search-setup/SKILL.md) for the upgrade
  140. steps.
  141. The CLI-backed skills are separate: deep research, enrichment, FindAll, and
  142. Monitor run through `parallel-cli` against your own funded Parallel account and
  143. are billed as ordinary API usage.
  144. ## Privacy and terms
  145. Parallel hosts the MCP service. Tool arguments sent to Parallel can include
  146. objectives, search queries, requested URLs, research questions, decision inputs,
  147. and optional session and model metadata. Parallel also receives standard
  148. connection and MCP client metadata.
  149. This data is handled under the
  150. [Customer Terms](https://parallel.ai/customer-terms) and
  151. [Privacy Policy](https://parallel.ai/privacy-policy). The MCP server does not
  152. automatically receive the rest of your conversation or local files, but an agent
  153. can include context in tool arguments when it is relevant. Do not send secrets,
  154. credentials, or sensitive local content through MCP tools.
  155. The CLI-backed skills run `parallel-cli` locally using credentials you have
  156. already configured. They send task inputs to Parallel's API under the same terms.
  157. Claude uses OAuth when account access is needed. Codex and
  158. Antigravity retain anonymous MCP connections. No bundled connection forwards
  159. CLI credentials. CLI-backed skills authenticate separately to Parallel through
  160. `parallel-cli`; the setup skill can check authentication status and start the
  161. CLI's sign-in flow.
  162. Migration examples describe authentication for the application being migrated.
  163. Provider credentials shown in scanner tests are synthetic source-code fixtures,
  164. not real keys or live requests. Neither those examples nor the test fixtures run
  165. automatically when the plugin loads.
  166. ## Support and security
  167. - **Questions and bugs:** [open an issue](https://github.com/parallel-web/parallel-agent-skills/issues) with your client, version, and error message (leave out API keys and other credentials), or see the [documentation](https://docs.parallel.ai/home).
  168. - **Product support:** contact Parallel at [support@parallel.ai](mailto:support@parallel.ai).
  169. - **Security concerns:** report privately to [support@parallel.ai](mailto:support@parallel.ai) — see [SECURITY.md](SECURITY.md). Please don't file security reports as public issues.
  170. ## Contributing
  171. See [MAINTAINERS.md](MAINTAINERS.md) for maintainer workflows, release process, and dev setup.
  172. ## Resources
  173. - [Documentation](https://docs.parallel.ai/home)
  174. - [API Platform](https://platform.parallel.ai)
  175. - [parallel-cli](https://github.com/parallel-web/parallel-web-tools)
  176. - [Pricing](https://parallel.ai/pricing)
  177. ## Local Development
  178. **Claude Code:**
  179. ```bash
  180. git clone https://github.com/parallel-web/parallel-agent-skills.git
  181. claude --plugin-dir /path/to/parallel-agent-skills
  182. /parallel:parallel-cli-setup
  183. ```
  184. **Codex:**
  185. ```bash
  186. git clone https://github.com/parallel-web/parallel-agent-skills.git
  187. cd parallel-agent-skills
  188. codex
  189. # Skills are auto-discovered via .agents/skills/
  190. ```
  191. ## License
  192. MIT

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.