Skip to content
Back to search

skills

Agent Skills for competitive GTM work - battlecards, win-loss, positioning, pricing, launches, deal prep and sales enablement. Installs into Claude Code, Codex, Copilot CLI, Cursor, Gemini CLI and any Agent Skills client. Works standalone; uses live evidence when LMTY is connected.

Safety scan

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

Checked today
Badge for your README
findsafeskills safety rating

Markdown

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

HTML

<a href="/listings/lmtydotcom--skills"><img src="/api/listings/lmtydotcom--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.

Each install below is the LMTY plugin: the full skill pack plus the LMTY connection. Start with the tool you already use.

ChatGPT and Codex

In the ChatGPT desktop app, the LMTY marketplace appears as a source in the Plugins Directory once it has been added; install the plugin from there and it is available in Codex and in ChatGPT Work.

To add the marketplace, from the Codex CLI:

codex plugin marketplace add LMTYdotcom/skills
codex plugin add lmty@lmty-plugins

Skills are invoked as `$competitive-battlecard` in Codex and with `@` in ChatGPT.

Claude Code

Inside a Claude Code session, type `/plugin marketplace add LMTYdotcom/skills`, then `/plugin install lmty@lmty-plugins`.

Or run these two commands in your terminal:

claude plugin marketplace add LMTYdotcom/skills
claude plugin install lmty@lmty-plugins

To get a newer version later:

claude plugin marketplace update lmty-plugins
claude plugin update lmty@lmty-plugins

To try it without installing, download the release zip and run `claude --plugin-dir lmty-<version>.zip`.

More tools and the command line

<details> <summary>GitHub Copilot, Cursor, Gemini CLI, and any other assistant</summary>

GitHub Copilot CLI

copilot plugin marketplace add LMTYdotcom/skills
copilot plugin install lmty@lmty-plugins

Or install straight from the repository, without a marketplace:

copilot plugin install LMTYdotcom/skills

Cursor

Cursor loads the pack as an Agent Plugin. Clone or unzip it into `~/.cursor/plugins/local/lmty`, restart Cursor, and the skills appear under **Customize**. Teams and Enterprise workspaces can add the repository as a team marketplace instead, from **Dashboard -> Plugins**.

Gemini CLI

gemini extensions install https://github.com/LMTYdotcom/skills

OpenCode, Antigravity, Windsurf, Amp, Kiro, Cline and the rest

The [skills CLI](https://github.com/vercel-labs/skills) installs the skills into whichever assistants it finds on your machine:

npx skills add LMTYdotcom/skills

That asks which skills and which assistants you want. To take the whole pack into every assistant it finds without being asked, add `--all`, which is shorthand for `--skill '*' --agent '*' -y`:

npx skills add LMTYdotcom/skills --all

To install just one skill, add `--skill` and its name:

npx skills add LMTYdotcom/skills --skill competitive-battlecard

This route installs the skills only. To also connect LMTY, add the MCP server to your assistant by hand; see [Connect LMTY](#connect-lmty).

Download

Every release on the [releases page](https://github.com/LMTYdotcom/skills/releases/latest) includes `lmty-<version>.zip`. Unzip it and point your assistant at the folder, or copy a single folder out of `skills/` into wherever your assistant keeps skills (for example `~/.claude/skills/`). Each skill is self-contained.

Releases are immutable and signed. To check that a zip is the one this repository published:

gh release verify-asset v1.0.0 lmty-1.0.0.zip -R LMTYdotcom/skills

</details>

<details> <summary>Versions and updates</summary>

Most installs track the `main` branch, not a release, and every client decides for itself when to pull a change:

| Client | Installs from | Updates when | | --- | --- | --- | | Claude Code | `main` | you run `claude plugin update`, and the version has changed | | ChatGPT and Codex | `main` | you refresh the marketplace | | GitHub Copilot CLI | `main` | you run `copilot plugin update` | | Cursor | the branch the marketplace tracks | that branch moves | | Gemini CLI | the latest release | a newer release is published | | skills CLI | `main` | you run `npx skills update` |

`main` is always a releasable state - every push is validated - so tracking it is the intended default.

To pin to a release instead, name it at install time:

npx skills add LMTYdotcom/skills#v1.0.0

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 · 49 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. {
  2. "$schema": "https://json.schemastore.org/claude-code-marketplace.json",
  3. "name": "lmty-plugins",
  4. "description": "Plugins by LMTY.",
  5. "owner": {
  6. "name": "LMTY",
  7. "email": "contact@lmty.com",
  8. "url": "https://lmty.com"
  9. },
  10. "plugins": [
  11. {
  12. "name": "lmty",
  13. "displayName": "Competitive Skills by LMTY",
  14. "source": "./",
  15. "description": "AI skills for competitive GTM work across market intelligence, positioning, pricing, launches, sales enablement, competitive deals, win/loss, messaging and product planning. Works standalone and uses current competitors, market history and evidence when LMTY is connected.",
  16. "category": "marketing",
  17. "author": {
  18. "name": "LMTY",
  19. "email": "contact@lmty.com",
  20. "url": "https://lmty.com"
  21. },
  22. "homepage": "https://lmty.com",
  23. "repository": "https://github.com/LMTYdotcom/skills",
  24. "license": "MIT",
  25. "keywords": [
  26. "competitive",
  27. "competitive-intelligence",
  28. "market-intelligence",
  29. "gtm",
  30. "go-to-market",
  31. "pmm",
  32. "product-marketing",
  33. "competitor-analysis",
  34. "positioning",
  35. "pricing",
  36. "battlecard",
  37. "launch",
  38. "sales-enablement",
  39. "sales",
  40. "objection-handling",
  41. "deal-prep",
  42. "win-loss",
  43. "messaging",
  44. "roadmap",
  45. "lmty"
  46. ]
  47. }
  48. ]
  49. }
README.md · README · 298 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # Competitive Skills by LMTY
  2. [![Installs](https://skills.sh/b/LMTYdotcom/skills)](https://skills.sh/LMTYdotcom/skills)
  3. [![Validate](https://github.com/LMTYdotcom/skills/actions/workflows/validate.yaml/badge.svg)](https://github.com/LMTYdotcom/skills/actions/workflows/validate.yaml)
  4. [![Release](https://img.shields.io/github/v/release/LMTYdotcom/skills?label=release)](https://github.com/LMTYdotcom/skills/releases/latest)
  5. [![OpenSSF Best Practices](https://www.bestpractices.dev/projects/14999/badge)](https://www.bestpractices.dev/projects/14999)
  6. [![License](https://img.shields.io/badge/license-MIT-blue.svg)](LICENSE)
  7. Use your AI assistant for competitive work you already do: researching
  8. competitors and market shifts, building positioning and messaging, reviewing
  9. pricing and launches, creating battlecards, preparing deals, analyzing wins
  10. and losses, and keeping sales enablement current.
  11. Add the pack once, then ask in plain English, "Give me this week's
  12. competitive briefing." Your assistant selects the matching skill and follows
  13. a repeatable method for completing the work. The skills work in Claude Code,
  14. ChatGPT and Codex, GitHub Copilot, Cursor, Gemini CLI and any assistant that
  15. supports [Agent Skills](https://agentskills.io); the pack itself is an
  16. [Agent Plugin](https://agent-plugins.org).
  17. Every skill works on its own. Connect [LMTY](https://lmty.com) and each one
  18. starts from your tracked competitors, kept current with what has changed and
  19. the evidence behind it, instead of a blank page.
  20. ## What you can ask
  21. | Skill | Ask something like | You get |
  22. | --- | --- | --- |
  23. | [`competitive-market-briefing`](skills/competitive-market-briefing/SKILL.md) | "Give me this week's competitive briefing." | A calm digest of what actually changed, and what to watch |
  24. | [`competitive-deep-dive`](skills/competitive-deep-dive/SKILL.md) | "Get me fully up to speed on Northwind." | One competitor, understood and evidenced |
  25. | [`competitive-market-shift-analysis`](skills/competitive-market-shift-analysis/SKILL.md) | "Everyone is adding AI. Is this a real shift or just noise?" | A verdict: noise, one move, an emerging shift, or a new expectation |
  26. | [`competitive-executive-market-brief`](skills/competitive-executive-market-brief/SKILL.md) | "Board meeting Friday. What market changes could blindside me?" | What leadership must decide, with the uncertainty shown |
  27. | [`competitive-positioning`](skills/competitive-positioning/SKILL.md) | "Our competitors all sound like us now. Pressure-test our positioning." | A position tested against the alternatives buyers really consider |
  28. | [`competitive-messaging-audit`](skills/competitive-messaging-audit/SKILL.md) | "Does our homepage still differentiate us?" | What still works, what went generic, what buyers would say instead |
  29. | [`competitive-pricing-packaging-review`](skills/competitive-pricing-packaging-review/SKILL.md) | "Northwind raised prices. What should we look at before changing ours?" | The market evidence a pricing decision needs, and the decision named |
  30. | [`competitive-launch-readiness`](skills/competitive-launch-readiness/SKILL.md) | "Here's the PRD. What do I need to know before we launch?" | Whether it's parity, catch-up or real differentiation, and how to frame it |
  31. | [`competitive-battlecard`](skills/competitive-battlecard/SKILL.md) | "Build a battlecard for us vs Northwind." | A card a rep can use with two minutes before a call |
  32. | [`competitive-objection-response`](skills/competitive-objection-response/SKILL.md) | "The buyer just said Northwind is cheaper. What do I say?" | The facts checked, a response, and the question to ask next |
  33. | [`competitive-deal-prep`](skills/competitive-deal-prep/SKILL.md) | "Prep me for the Globex call tomorrow - we're up against Northwind." | One deal, prepared for the next meeting |
  34. | [`competitive-poc-rfp-plan`](skills/competitive-poc-rfp-plan/SKILL.md) | "We're going into a POC against Northwind. What should we prove?" | An evaluation planned around the criteria that will decide it |
  35. | [`competitive-enablement-refresh`](skills/competitive-enablement-refresh/SKILL.md) | "Northwind changed their pricing. What in our talk tracks is now wrong?" | The lines that changed, with proof and date |
  36. | [`competitive-asset-audit`](skills/competitive-asset-audit/SKILL.md) | "Audit our competitive materials. Which claims are stale?" | Every competitive claim, marked and ranked by exposure |
  37. | [`competitive-win-loss-analysis`](skills/competitive-win-loss-analysis/SKILL.md) | "Why are we losing to Northwind?" | What buyers actually said, counted, with the exceptions kept |
  38. | [`competitive-roadmap-review`](skills/competitive-roadmap-review/SKILL.md) | "Competitors keep adding this. Does it matter for our roadmap?" | Product questions from market and customer evidence, not a copied feature list |
  39. You do not pick a skill; the right one is chosen for you. If a request needs
  40. more than one, the assistant does the part that fits and tells you which
  41. skill to run next.
  42. You can also call a skill directly: `/lmty:competitive-battlecard Northwind`
  43. in Claude Code and GitHub Copilot, `$competitive-battlecard` in Codex, `@`
  44. in ChatGPT.
  45. ## Connect LMTY
  46. Without LMTY, a skill researches the competitor from scratch each time and
  47. tells you the answer is a snapshot. With LMTY connected, it already knows
  48. your product, your tracked competitors, what changed and when, and the
  49. sources behind each claim.
  50. | Without LMTY | With LMTY |
  51. | --- | --- |
  52. | Asks who "we" are and which competitor | Already knows your product and your competitor set |
  53. | Researches a snapshot, dated today | Reads tracked market state with a history of changes |
  54. | "Not observed" | "Not observed, last checked on `<date>`" |
  55. | Sources are whatever it found this session | Every finding links back to its public source, with when LMTY observed it. |
  56. [LMTY](https://lmty.com) tracks a product's competitors and keeps a current
  57. report plus a log of what moved.
  58. The connection is already set up for you. This pack ships with the LMTY
  59. server configured, so in Claude Code, ChatGPT and Codex, GitHub Copilot CLI,
  60. Cursor and Gemini CLI your assistant will ask you to sign in, in your
  61. browser - ChatGPT and Codex as part of installing, the others the first time
  62. you open a session with the plugin enabled. Approve it once and every skill
  63. here uses it from then on.
  64. No skill needs LMTY to run, and none will refuse work without it. If you
  65. never connect, the skills research from scratch and say so.
  66. Connecting sends your requests to LMTY's server. LMTY's
  67. [Privacy policy](https://lmty.com/privacy) and
  68. [Terms of service](https://lmty.com/terms) cover what happens to them.
  69. <details>
  70. <summary>Connecting by hand, or with an API token</summary>
  71. If your assistant installed the skills through `npx skills add`, it did not
  72. see the server. Add it to your assistant's MCP configuration as a Streamable
  73. HTTP server at `https://mcp.lmty.com/mcp`; sign-in is discovered
  74. automatically.
  75. If you would rather not use the browser sign-in, LMTY can issue a token from
  76. your account settings. Add the server to your own configuration with an
  77. `Authorization: Bearer <token>` header rather than relying on the bundled
  78. one.
  79. </details>
  80. ## Install
  81. Each install below is the LMTY plugin: the full skill pack plus the LMTY
  82. connection. Start with the tool you already use.
  83. ### ChatGPT and Codex
  84. In the ChatGPT desktop app, the LMTY marketplace appears as a source in the
  85. Plugins Directory once it has been added; install the plugin from there and
  86. it is available in Codex and in ChatGPT Work.
  87. To add the marketplace, from the Codex CLI:
  88. ```bash
  89. codex plugin marketplace add LMTYdotcom/skills
  90. codex plugin add lmty@lmty-plugins
  91. ```
  92. Skills are invoked as `$competitive-battlecard` in Codex and with `@` in
  93. ChatGPT.
  94. ### Claude Code
  95. Inside a Claude Code session, type `/plugin marketplace add
  96. LMTYdotcom/skills`, then `/plugin install lmty@lmty-plugins`.
  97. Or run these two commands in your terminal:
  98. ```bash
  99. claude plugin marketplace add LMTYdotcom/skills
  100. claude plugin install lmty@lmty-plugins
  101. ```
  102. To get a newer version later:
  103. ```bash
  104. claude plugin marketplace update lmty-plugins
  105. claude plugin update lmty@lmty-plugins
  106. ```
  107. To try it without installing, download the release zip and run
  108. `claude --plugin-dir lmty-<version>.zip`.
  109. ### More tools and the command line
  110. <details>
  111. <summary>GitHub Copilot, Cursor, Gemini CLI, and any other assistant</summary>
  112. #### GitHub Copilot CLI
  113. ```bash
  114. copilot plugin marketplace add LMTYdotcom/skills
  115. copilot plugin install lmty@lmty-plugins
  116. ```
  117. Or install straight from the repository, without a marketplace:
  118. ```bash
  119. copilot plugin install LMTYdotcom/skills
  120. ```
  121. #### Cursor
  122. Cursor loads the pack as an Agent Plugin. Clone or unzip it into
  123. `~/.cursor/plugins/local/lmty`, restart Cursor, and the skills appear under
  124. **Customize**. Teams and Enterprise workspaces can add the repository as a
  125. team marketplace instead, from **Dashboard -> Plugins**.
  126. #### Gemini CLI
  127. ```bash
  128. gemini extensions install https://github.com/LMTYdotcom/skills
  129. ```
  130. #### OpenCode, Antigravity, Windsurf, Amp, Kiro, Cline and the rest
  131. The [skills CLI](https://github.com/vercel-labs/skills) installs the skills
  132. into whichever assistants it finds on your machine:
  133. ```bash
  134. npx skills add LMTYdotcom/skills
  135. ```
  136. That asks which skills and which assistants you want. To take the whole pack
  137. into every assistant it finds without being asked, add `--all`, which is
  138. shorthand for `--skill '*' --agent '*' -y`:
  139. ```bash
  140. npx skills add LMTYdotcom/skills --all
  141. ```
  142. To install just one skill, add `--skill` and its name:
  143. ```bash
  144. npx skills add LMTYdotcom/skills --skill competitive-battlecard
  145. ```
  146. This route installs the skills only. To also connect LMTY, add the MCP
  147. server to your assistant by hand; see [Connect LMTY](#connect-lmty).
  148. #### Download
  149. Every release on the [releases page](https://github.com/LMTYdotcom/skills/releases/latest)
  150. includes `lmty-<version>.zip`. Unzip it and point your assistant at the
  151. folder, or copy a single folder out of `skills/` into wherever your
  152. assistant keeps skills (for example `~/.claude/skills/`). Each skill is
  153. self-contained.
  154. Releases are immutable and signed. To check that a zip is the one this
  155. repository published:
  156. ```bash
  157. gh release verify-asset v1.0.0 lmty-1.0.0.zip -R LMTYdotcom/skills
  158. ```
  159. </details>
  160. <details>
  161. <summary>Versions and updates</summary>
  162. Most installs track the `main` branch, not a release, and every client
  163. decides for itself when to pull a change:
  164. | Client | Installs from | Updates when |
  165. | --- | --- | --- |
  166. | Claude Code | `main` | you run `claude plugin update`, and the version has changed |
  167. | ChatGPT and Codex | `main` | you refresh the marketplace |
  168. | GitHub Copilot CLI | `main` | you run `copilot plugin update` |
  169. | Cursor | the branch the marketplace tracks | that branch moves |
  170. | Gemini CLI | the latest release | a newer release is published |
  171. | skills CLI | `main` | you run `npx skills update` |
  172. `main` is always a releasable state - every push is validated - so tracking
  173. it is the intended default.
  174. To pin to a release instead, name it at install time:
  175. ```bash
  176. npx skills add LMTYdotcom/skills#v1.0.0
  177. gemini extensions install https://github.com/LMTYdotcom/skills --ref=v1.0.0
  178. codex plugin marketplace add LMTYdotcom/skills --ref v1.0.0
  179. ```
  180. In the skills CLI, `#` takes a tag, branch or commit, while `@` picks a
  181. skill, so `LMTYdotcom/skills#v1.0.0@competitive-battlecard` is one skill
  182. at one version. `npx skills update` then keeps re-checking the ref you
  183. pinned rather than moving you forward.
  184. </details>
  185. ## How the skills behave
  186. Competitive work goes wrong in a few repeatable ways. Most of what is in
  187. these skills exists to stop one of them.
  188. - **Absence is not evidence.** A pricing page that does not mention audit
  189. logging says something about the page, not the product. The skills write
  190. "not observed in the sources checked, as of a date," never "does not
  191. have."
  192. - **No invented motive.** A competitor raised prices; why is unknowable
  193. without a statement from them, and the skills say so instead of guessing.
  194. - **Dates travel with claims.** Every material fact carries the date it was
  195. observed, and coverage is tracked separately from freshness.
  196. - **Fact and interpretation are kept apart.** Observed fact, interpretation,
  197. implication, and action are labeled, and any ranking is marked as the
  198. skill's own judgment.
  199. - **Recommendations name the decision, not the answer.** Without your
  200. internal numbers and deal evidence, the useful output is the question and
  201. the evidence needed to settle it.
  202. - **Empty sections stay empty.** "No validated objections yet - pull the
  203. last 20 calls where this competitor was named" is more useful than three
  204. plausible ones.
  205. - **They hand off.** A request that straddles two skills gets the half this
  206. skill owns and the name of the one that owns the rest.
  207. ## How they were tested
  208. Every skill was run against three models from different vendors, with a
  209. fourth model grading the output against written criteria, and then run live
  210. in Claude Code against the real LMTY server. Across the pack, the right
  211. skill is chosen for 95% of example requests, including the ones written to
  212. sit on the boundary between two neighbors.
  213. The test criteria are kept out of this repository on purpose: a skill should
  214. not ship with its own answer key.
  215. ## Contributing
  216. Bug reports and suggestions are welcome as
  217. [issues](https://github.com/LMTYdotcom/skills/issues). If you want to change
  218. a skill, [`CONTRIBUTING.md`](CONTRIBUTING.md) explains how the repository is
  219. laid out and what a change needs before it merges. Taking part means
  220. following the [Code of Conduct](CODE_OF_CONDUCT.md).
  221. Need help rather than a change? [`SUPPORT.md`](SUPPORT.md) says where to ask.
  222. ## License
  223. MIT. See [`LICENSE`](LICENSE).

Ratings and comments

Sign in to rate this skill.

Sign in

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