Skip to content
Back to search

synalinks-skills

Coding Agents skills for Synalinks OSS

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 13 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/SynaLinks/synalinks-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/SynaLinks/synalinks-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/synalinks--synalinks-skills/badge.svg)](/listings/synalinks--synalinks-skills)

HTML

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

Install it as a plugin in one command:

# Claude Code (2.1.275 or later)
/plugin install synalinks --marketplace SynaLinks/synalinks-skills

# Codex
codex plugin marketplace add SynaLinks/synalinks-skills
codex plugin add synalinks@synalinks-skills

# OpenCode
opencode plugin github:SynaLinks/synalinks-skills

Any other agent (Cursor, Gemini CLI, Copilot, pi, ...):

npx skills add SynaLinks/synalinks-skills --skill synalinks

Details, global installs and Claude.ai uploads are in [Install](#install).

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 · 8 lines
  1. {
  2. "name": "synalinks-skills",
  3. "owner": { "name": "SynaLinks" },
  4. "description": "Synalinks skills for coding agents",
  5. "plugins": [
  6. { "name": "synalinks", "source": "./", "description": "Idiomatic Synalinks for coding agents" }
  7. ]
  8. }
README.md · README · 285 lines
  1. # Agent Skills for Synalinks
  2. [![CI](https://github.com/SynaLinks/synalinks-skills/actions/workflows/ci.yml/badge.svg)](https://github.com/SynaLinks/synalinks-skills/actions/workflows/ci.yml)
  3. ---
  4. This repository contains skills for coding agents that read the open Agent
  5. Skills format (`SKILL.md`): Claude Code, Codex, OpenCode, pi, and others.
  6. ## Quick install
  7. Install it as a plugin in one command:
  8. ```shell
  9. # Claude Code (2.1.275 or later)
  10. /plugin install synalinks --marketplace SynaLinks/synalinks-skills
  11. # Codex
  12. codex plugin marketplace add SynaLinks/synalinks-skills
  13. codex plugin add synalinks@synalinks-skills
  14. # OpenCode
  15. opencode plugin github:SynaLinks/synalinks-skills
  16. ```
  17. Any other agent (Cursor, Gemini CLI, Copilot, pi, ...):
  18. ```shell
  19. npx skills add SynaLinks/synalinks-skills --skill synalinks
  20. ```
  21. Details, global installs and Claude.ai uploads are in [Install](#install).
  22. ## What are Agent Skills?
  23. Agent Skills are reusable instructions, in a plain `SKILL.md` file, that teach a
  24. coding agent how to perform a task the way you want it done. The format started
  25. at Anthropic and is now an open standard read by agents across the ecosystem, so
  26. one skill works in every agent that supports it, unmodified.
  27. ## What is the goal of Synalinks skills?
  28. The goal of Synalinks skills is to teach coding agents to use the Synalinks
  29. framework correctly. Synalinks is Keras-inspired, so without guidance LMs tend to
  30. mix Keras / LangChain / DSPy syntax, producing plausible-looking but broken code.
  31. These skills constrain the agent to idiomatic Synalinks usage.
  32. ## The `synalinks` skill
  33. This repo ships a **single skill**, `synalinks`, that covers the whole
  34. framework. The agent auto-activates it based on its `SKILL.md` `description`. The
  35. `SKILL.md` is a scannable overview, organized into one section per slice of the
  36. framework, each pointing to a runnable script (and its captured run log) and a
  37. deep-dive reference doc:
  38. | Section | What it covers |
  39. |---------|----------------|
  40. | **Core** | DataModel, Field, Input, JSON operators (`+ & \| ^ ~`), `synalinks.ops`, configuration, LanguageModel/EmbeddingModel basics, DecisionModel |
  41. | **Programs** | Program class, four building APIs (Functional / Sequential / Subclassing / Mixed), multi-input/output graphs, build/call lifecycle, save/load, `summary`, `get_module`, `plot_program`, custom serialization |
  42. | **Modules** | Generator, ChainOfThought, SelfCritique, Identity, PythonSynthesis, SequentialPlanSynthesis, custom modules via subclassing |
  43. | **Control flow** | Decision, MultiDecision, Branch, parallel branches, self-consistency, XOR input/output guards, And/Or modules, branch merging |
  44. | **Decision models** | DecisionModel (TypeSafe jev), fields as questions, the `decision_model` argument, `min_confidence`/`threshold` abstention, tuning thresholds with KerasTuner |
  45. | **Agents** | FunctionCallingAgent, RLM, DeepAgent, Tool definitions, MCP integration (MultiServerMCPClient), subagents, trajectories |
  46. | **Knowledge** | KnowledgeBase (DuckDB), EmbedKnowledge, UpdateKnowledge, RetrieveKnowledge, RAG/KAG, hybrid search, Entity/Relation graphs |
  47. | **Training** | `compile()` / `fit()` / `evaluate()` / `predict()`, callbacks, ProgramCheckpoint, training workflow |
  48. | **Rewards** | ExactMatch, CosineSimilarity, LMAsJudge, ProgramAsJudge, RubricsAsJudge and its presets (Faithfulness, Toxicity, TaskCompletion, ...), AgentAsJudge, DeepAgentAsJudge, RLMAsJudge, ComposableReward, BatchReward, F1Score, custom rewards/metrics, masking |
  49. | **Optimizers** | RandomFewShot, OMEGA, Dominated Novelty Search, mutation/crossover, quality-diversity tuning |
  50. | **Providers** | Provider prefixes (openai, anthropic, groq, openrouter, cohere, deepseek, together_ai, bedrock, doubleword, hosted_vllm, ...), local OpenAI-compatible servers (LMStudio/vLLM/ollama), OpenAI-compatible setups (e.g. OpenRouter embeddings via `api_base`) |
  51. | **Datasets** | Built-in datasets (gsm8k, hotpotqa, arcagi), custom iterable datasets, visualization (`plot_program`, `plot_history`, `plot_metrics_*`) |
  52. The skill folder contains:
  53. - `SKILL.md`: frontmatter + scannable usage guide for the whole framework
  54. - `scripts/`: runnable example scripts (each docstring states its `Usage:` and `Run log:`)
  55. - `references/`: deep-dive reference docs plus the captured run logs (`*.log`)
  56. ## Install
  57. Pick the route that matches your agent:
  58. | Agent | Install with |
  59. | --- | --- |
  60. | Claude Code, Codex, OpenCode | [the plugin](#as-a-plugin) (recommended) |
  61. | Any other agent (Cursor, Gemini CLI, Copilot, pi, ...) | [`npx skills`](#with-npx-skills) |
  62. | Claude.ai (web) and Claude Desktop | [a `.skill` upload](#claudeai-web-and-claude-desktop) |
  63. The skill content is identical in every case: one `SKILL.md`, no per-agent
  64. variants to maintain.
  65. ### As a plugin
  66. This repo is a plugin marketplace for Claude Code and Codex, and an OpenCode
  67. plugin. The agent fetches the skill itself: no Node.js, no copied files.
  68. **Claude Code** (2.1.275 or later):
  69. ```shell
  70. /plugin install synalinks --marketplace SynaLinks/synalinks-skills
  71. ```
  72. Then run `/skills`: `synalinks` should appear.
  73. **Codex:**
  74. ```shell
  75. codex plugin marketplace add SynaLinks/synalinks-skills
  76. codex plugin add synalinks@synalinks-skills
  77. ```
  78. Start a new Codex session to load it.
  79. **OpenCode:**
  80. ```shell
  81. opencode plugin github:SynaLinks/synalinks-skills # this project
  82. opencode plugin github:SynaLinks/synalinks-skills --global # every project
  83. ```
  84. This adds the plugin to `.opencode/opencode.json` (or your global config). You
  85. can also list it yourself:
  86. ```json
  87. {
  88. "$schema": "https://opencode.ai/config.json",
  89. "plugin": ["github:SynaLinks/synalinks-skills"]
  90. }
  91. ```
  92. The plugin registers the repo's `skills/` directory as a skill source, so
  93. OpenCode discovers the skill without copying files. Restart OpenCode to load it,
  94. then check with `opencode debug skill`.
  95. ### With `npx skills`
  96. For any other agent, use the official [`skills`](https://skills.sh) CLI
  97. (`npx skills`). It clones this repo, discovers the `synalinks` skill, and wires
  98. it into your agent's skills directory: no manual cloning, copying, or zipping
  99. required. It needs Node.js (the `npx` that ships with it); nothing else. It also
  100. works for Claude Code, Codex and OpenCode if you prefer it to the plugin.
  101. `synalinks` is a skill for the **synalinks Python library**, so the natural place
  102. to install it is **the project where you use synalinks**, so the agent
  103. gets idiomatic-synalinks guidance exactly where you write synalinks code. Project
  104. scope is also the CLI default, so just run this from your project root:
  105. ```shell
  106. npx skills add SynaLinks/synalinks-skills --skill synalinks
  107. ```
  108. The CLI detects the agents installed on your machine and installs into the
  109. current project (each agent reads it from its own directory, e.g.
  110. `<repo>/.cursor/skills/` or `<repo>/.agents/skills/`). Commit the generated
  111. `skills-lock.json` (see [Updating](#updating)) so teammates and CI resolve the
  112. same skill content.
  113. Want it in **every** project instead of per-project? Install globally:
  114. ```shell
  115. npx skills add SynaLinks/synalinks-skills --skill synalinks --global
  116. ```
  117. That installs to the agent's user-level skills directory. By default the CLI
  118. symlinks the skill; pass `--copy` to copy the files instead.
  119. Target one agent (or all) explicitly with `-a`/`--agent`:
  120. ```shell
  121. npx skills add SynaLinks/synalinks-skills --skill synalinks -a cursor # Cursor
  122. npx skills add SynaLinks/synalinks-skills --skill synalinks -a gemini-cli # Gemini CLI
  123. npx skills add SynaLinks/synalinks-skills --skill synalinks -a pi # pi
  124. npx skills add SynaLinks/synalinks-skills --all # everything, all agents
  125. ```
  126. The CLI supports 70+ agents. Passing an unknown name prints the full list of
  127. valid agent slugs, which currently includes claude-code, codex, opencode,
  128. github-copilot, windsurf, zed, crush, goose, junie and roo, among others.
  129. List what's in the repo without installing anything:
  130. ```shell
  131. npx skills add SynaLinks/synalinks-skills --list
  132. ```
  133. Run `npx skills add SynaLinks/synalinks-skills` with no flags for an
  134. interactive picker, and `npx skills list` to see what is installed.
  135. Agents auto-activate the skill for your task based on the `SKILL.md`
  136. `description` field, so there is nothing to select manually.
  137. ### Claude.ai (web) and Claude Desktop
  138. Web/Desktop uploads expect a single `.skill` archive (a zip of the skill folder
  139. with `SKILL.md` at the root). Clone the repo and build the archive:
  140. ```shell
  141. git clone https://github.com/SynaLinks/synalinks-skills.git
  142. cd synalinks-skills/skills
  143. zip -r ../synalinks.skill synalinks
  144. ```
  145. Then in the Claude interface, click the skill icon (🧩), choose "Upload
  146. skill", and upload the `.skill` file. See [Using skills with Claude](https://support.claude.com/en/articles/12512180-using-skills-in-claude#h_c6008b84ad)
  147. for the latest UI walkthrough.
  148. ## Updating
  149. The `synalinks` skill is **not** semver-versioned: it tracks the `main` branch
  150. of this repo. "Latest" always means the current `main` (the CLI installs from
  151. the default branch's HEAD; there is no version tag to pin). Update with:
  152. ```shell
  153. npx skills update # update every installed skill to latest
  154. npx skills update synalinks # just this skill
  155. ```
  156. Use `-g`/`--global` or `-p`/`--project` to scope the update. The CLI re-fetches
  157. this repo, recomputes the skill's content hash, and re-installs only if it
  158. changed from what's recorded.
  159. ### Lockfile (reproducible installs)
  160. Installing writes a `skills-lock.json` in your project that pins each skill to a
  161. source repo + a SHA-256 content hash, e.g.:
  162. ```json
  163. {
  164. "version": 1,
  165. "skills": {
  166. "synalinks": {
  167. "source": "SynaLinks/synalinks-skills",
  168. "sourceType": "github",
  169. "skillPath": "skills/synalinks/SKILL.md",
  170. "computedHash": "<sha256-of-the-skill>"
  171. }
  172. }
  173. }
  174. ```
  175. Commit `skills-lock.json` so teammates get the exact same skill content.
  176. `npx skills update` advances the recorded hash to the latest `main`; to restore
  177. the pinned state instead of updating, use:
  178. ```shell
  179. npx skills experimental_install # reinstall skills from skills-lock.json
  180. ```
  181. For Claude.ai / Desktop (which take a `.skill` archive rather than the CLI),
  182. regenerate the archive and re-upload it:
  183. ```shell
  184. cd synalinks-skills/skills && zip -r ../synalinks.skill synalinks
  185. ```
  186. ## Uninstall
  187. ```shell
  188. npx skills remove synalinks
  189. ```
  190. Add `-g`/`--global` to remove a globally-installed skill. For Claude.ai /
  191. Desktop, remove the skill from the skill icon (🧩) menu.
  192. # License
  193. These skills are licensed under Apache 2.0, like Synalinks framework.
  194. See the [LICENSE](LICENSE) file for full details.
  195. ## Contributing
  196. Contributions are welcome! Please feel free to submit a Pull Request.
  197. Every push and pull request runs the [CI workflow](.github/workflows/ci.yml),
  198. which validates each `skills/*/SKILL.md` against the Agent Skills spec and
  199. installs the skills into every agent the `skills` CLI supports, so a broken
  200. frontmatter or a missing file is caught before it reaches users. On `main` it
  201. also re-installs from GitHub (the exact command users run) and uploads the
  202. `.skill` archives as a workflow artifact.
  203. ## Acknoledgement
  204. These skills have been created by [Ramiro Salas](https://www.linkedin.com/in/rsalas/) the CTO of [Hexatropian](https://www.linkedin.com/company/hextropian-systems/) an active early member of [Synalinks](https://github.com/SynaLinks/synalinks) community.
  205. ## Synalinks Project
  206. - [Synalinks Framework](https://github.com/SynaLinks/synalinks): the neuro-symbolic AI framework these skills are designed for

Ratings and comments

Sign in to rate this skill.

Sign in

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