Adoption

Agent Skills are supported by leading AI development tools.

VS Code Gemini CLI GitHub Goose Amp Cursor Claude Code Letta OpenCode Claude OpenAI Codex Factory VS Code Gemini CLI GitHub Goose Amp Cursor Claude Code Letta OpenCode Claude OpenAI Codex Factory

lifemate-ai/familiar-add-tool

Name: familiar-add-tool
Author: lifemate-ai

dev/familiar-add-tool/SKILL.md

npx skillsauth add lifemate-ai/familiar-ai familiar-add-tool

Clean

TrivyContainer and dependency vulnerability scanner

Clean

SemgrepStatic code analysis for vulnerabilities

Clean

mcp-scan (Snyk)Model Context Protocol security validation

Skipped

Snyk (dep)Open source security scanning

Skipped

Socket.devSupply chain security analysis

Skipped

VirusTotalMulti-engine malware detection

Skipped

CrowdStrikeAdvanced threat intelligence

Skipped

OSV-ScannerOpen Source Vulnerability database check

Skipped

OWASP Dep-Check

familiar-add-tool

Scaffold a new tool (sensor, actuator, or service) for the familiar-ai agent.

When to Use

Adding a new physical sensor (microphone, temperature, GPS, …)
Adding a new actuator (arm, LED, speaker, …)
Wrapping a new external service (weather API, home automation, …)

What This Skill Does

Creates src/familiar_agent/tools/<name>.py from the standard template
Registers the tool in agent.py in all three required places
Adds any new env vars to config.py and .env.example
Adds the tool's body description to the system prompt in agent.py

Step 1: Gather Requirements

Ask the user (or infer from context):

Tool name (snake_case, e.g. microphone)
Tool verbs — the callable actions the agent can take (e.g. listen, record)
Config vars — env vars needed (e.g. MIC_DEVICE_INDEX)
System prompt description — how should the agent understand this body part?
Optional flag — is the tool optional (guarded by config) or always available?

Step 2: Create the Tool File

Create src/familiar_agent/tools/<name>.py:

"""<Name> tool — <one-line description>."""

from __future__ import annotations

import logging

logger = logging.getLogger(__name__)


class <Name>Tool:
    """<Description>."""

    def __init__(self, <config_params>) -> None:
        self.<param> = <param>
        # TODO: initialize hardware/client

    async def <verb>(self, <args>) -> tuple[str, str | None]:
        """<What this action does>. Returns (text_result, image_b64_or_None)."""
        # TODO: implement
        return "Not implemented yet.", None

    def get_tool_definitions(self) -> list[dict]:
        return [
            {
                "name": "<verb>",
                "description": "<What the agent does when calling this tool>.",
                "input_schema": {
                    "type": "object",
                    "properties": {
                        # "<param>": {"type": "string", "description": "..."},
                    },
                    "required": [],
                },
            },
        ]

    async def call(self, tool_name: str, tool_input: dict) -> tuple[str, str | None]:
        if tool_name == "<verb>":
            return await self.<verb>(**tool_input)
        return f"Unknown tool: {tool_name}", None

Key rules for tool files:

call() always returns tuple[str, str | None] — text result + optional JPEG base64
get_tool_definitions() uses Anthropic input_schema format
Keep each tool in its own file, one class per file

Step 3: Register in `agent.py`

Touch three places:

3a. Import (top of file)

from .tools.<name> import <Name>Tool

3b. `_init_tools()` — instantiate if config is available

# In Agent._init_tools():
<name>_cfg = self.config.<name>          # or read directly from config
if <name>_cfg.<required_field>:
    self._<name> = <Name>Tool(<name>_cfg.<field>, ...)

3c. `_all_tool_defs` property — expose definitions

# In Agent._all_tool_defs:
if self._<name>:
    defs.extend(self._<name>.get_tool_definitions())

3d. `_execute_tool()` — route calls

# In Agent._execute_tool():
<name>_tools = {"<verb1>", "<verb2>"}

if name in <name>_tools and self._<name>:
    return await self._<name>.call(name, tool_input)

Also add self._<name>: <Name>Tool | None = None to __init__.

Step 4: Update `config.py`

If new env vars are needed, add a config dataclass:

@dataclass
class <Name>Config:
    <field>: str = field(default_factory=lambda: os.environ.get("<ENV_VAR>", ""))

And add it to AgentConfig:

<name>: <Name>Config = field(default_factory=<Name>Config)

Step 5: Update System Prompt in `agent.py`

Add the tool to the body parts description in SYSTEM_PROMPT:

- <Body part name> (<verb>): <How the agent should think about and use this>.

Keep it grounded and concrete — e.g. "Ears (listen): Your hearing. Calling listen() records audio from the microphone and transcribes it."

Step 6: Update `.env.example`

# <Name> tool
<ENV_VAR>=

Checklist Before Finishing

[ ] tools/<name>.py created with get_tool_definitions() and call()
[ ] Import added to agent.py
[ ] _init_tools() instantiates the tool (with optional guard)
[ ] _all_tool_defs includes it
[ ] _execute_tool() routes to it
[ ] config.py has new config class (if needed)
[ ] AgentConfig references the new config (if needed)
[ ] SYSTEM_PROMPT describes the new body part
[ ] .env.example documents new env vars
[ ] Run uv run ruff check . && uv run ruff format . before committing

lifemate-ai/familiar-add-tool

dev/familiar-add-tool/SKILL.md

Scaffold a new sensor/actuator tool for familiar-ai. Generates the tool file and registers it in all required places in agent.py.

75 stars

tools

Updated Apr 6, 2026

$ install --global

skillsauth

npx skillsauth add lifemate-ai/familiar-ai familiar-add-tool

Install this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.

Security Scan Results

3 of 9 scanners reported clean

Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.

Scanners Passed

Scanners in report

Clean

TrivyContainer and dependency vulnerability scanner

95%

Clean

SemgrepStatic code analysis for vulnerabilities

95%

Clean

mcp-scan (Snyk)Model Context Protocol security validation

95%

Skipped

Snyk (dep)Open source security scanning

50%

Skipped

Socket.devSupply chain security analysis

50%

Skipped

VirusTotalMulti-engine malware detection

50%

Skipped

CrowdStrikeAdvanced threat intelligence

50%

Skipped

OSV-ScannerOpen Source Vulnerability database check

50%

Skipped

OWASP Dep-Check

50%

Last scanned: Apr 6, 2026, 8:46 PM38.7s1 file scanned

SKILL.md

name:: familiar-add-tool
description:: Scaffold a new sensor/actuator tool for familiar-ai. Generates the tool file and registers it in all required places in agent.py.

familiar-add-tool

Scaffold a new tool (sensor, actuator, or service) for the familiar-ai agent.

When to Use

Adding a new physical sensor (microphone, temperature, GPS, …)
Adding a new actuator (arm, LED, speaker, …)
Wrapping a new external service (weather API, home automation, …)

What This Skill Does

Creates src/familiar_agent/tools/<name>.py from the standard template
Registers the tool in agent.py in all three required places
Adds any new env vars to config.py and .env.example
Adds the tool's body description to the system prompt in agent.py

Step 1: Gather Requirements

Ask the user (or infer from context):

Tool name (snake_case, e.g. microphone)
Tool verbs — the callable actions the agent can take (e.g. listen, record)
Config vars — env vars needed (e.g. MIC_DEVICE_INDEX)
System prompt description — how should the agent understand this body part?
Optional flag — is the tool optional (guarded by config) or always available?

Step 2: Create the Tool File

Create src/familiar_agent/tools/<name>.py:

"""<Name> tool — <one-line description>."""

from __future__ import annotations

import logging

logger = logging.getLogger(__name__)


class <Name>Tool:
    """<Description>."""

    def __init__(self, <config_params>) -> None:
        self.<param> = <param>
        # TODO: initialize hardware/client

    async def <verb>(self, <args>) -> tuple[str, str | None]:
        """<What this action does>. Returns (text_result, image_b64_or_None)."""
        # TODO: implement
        return "Not implemented yet.", None

    def get_tool_definitions(self) -> list[dict]:
        return [
            {
                "name": "<verb>",
                "description": "<What the agent does when calling this tool>.",
                "input_schema": {
                    "type": "object",
                    "properties": {
                        # "<param>": {"type": "string", "description": "..."},
                    },
                    "required": [],
                },
            },
        ]

    async def call(self, tool_name: str, tool_input: dict) -> tuple[str, str | None]:
        if tool_name == "<verb>":
            return await self.<verb>(**tool_input)
        return f"Unknown tool: {tool_name}", None

Key rules for tool files:

call() always returns tuple[str, str | None] — text result + optional JPEG base64
get_tool_definitions() uses Anthropic input_schema format
Keep each tool in its own file, one class per file

Step 3: Register in `agent.py`

Touch three places:

3a. Import (top of file)

from .tools.<name> import <Name>Tool

3b. `_init_tools()` — instantiate if config is available

# In Agent._init_tools():
<name>_cfg = self.config.<name>          # or read directly from config
if <name>_cfg.<required_field>:
    self._<name> = <Name>Tool(<name>_cfg.<field>, ...)

3c. `_all_tool_defs` property — expose definitions

# In Agent._all_tool_defs:
if self._<name>:
    defs.extend(self._<name>.get_tool_definitions())

3d. `_execute_tool()` — route calls

# In Agent._execute_tool():
<name>_tools = {"<verb1>", "<verb2>"}

if name in <name>_tools and self._<name>:
    return await self._<name>.call(name, tool_input)

Also add self._<name>: <Name>Tool | None = None to __init__.

Step 4: Update `config.py`

If new env vars are needed, add a config dataclass:

@dataclass
class <Name>Config:
    <field>: str = field(default_factory=lambda: os.environ.get("<ENV_VAR>", ""))

And add it to AgentConfig:

<name>: <Name>Config = field(default_factory=<Name>Config)

Step 5: Update System Prompt in `agent.py`

Add the tool to the body parts description in SYSTEM_PROMPT:

- <Body part name> (<verb>): <How the agent should think about and use this>.

Keep it grounded and concrete — e.g. "Ears (listen): Your hearing. Calling listen() records audio from the microphone and transcribes it."

Step 6: Update `.env.example`

# <Name> tool
<ENV_VAR>=

Checklist Before Finishing

[ ] tools/<name>.py created with get_tool_definitions() and call()
[ ] Import added to agent.py
[ ] _init_tools() instantiates the tool (with optional guard)
[ ] _all_tool_defs includes it
[ ] _execute_tool() routes to it
[ ] config.py has new config class (if needed)
[ ] AgentConfig references the new config (if needed)
[ ] SYSTEM_PROMPT describes the new body part
[ ] .env.example documents new env vars
[ ] Run uv run ruff check . && uv run ruff format . before committing

Related Skills

lifemate-ai/familiar-debug-loop

tools

VerifiedTrustedCommunity

Debug a stuck or misbehaving familiar-ai ReAct loop. Diagnoses tool-call failures, model output issues, prompt-mode parsing errors, and silent tool skips.

75SKILL.mdUpdated Apr 6, 2026

lifemate-ai/familiar-debug-loop

lifemate-ai/familiar-check-env

tools

VerifiedTrustedCommunity

Validate the .env configuration for familiar-ai. Detects common misconfigurations like wrong ports, missing MODEL, or TOOLS_MODE issues that cause silent failures or timeouts.

75SKILL.mdUpdated Apr 6, 2026

lifemate-ai/familiar-check-env

openclaw/taskflow

tools

VerifiedTrustedCommunity

Use when work should span one or more detached tasks but still behave like one job with a single owner context. TaskFlow is the durable flow substrate under authoring layers like Lobster, ACPX, plugins, or plain code. Keep conditional logic in the caller; use TaskFlow for flow identity, child-task linkage, waiting state, revision-checked mutations, and user-facing emergence.

357,764SKILL.mdUpdated Apr 10, 2026

openclaw/extensions/lobster

tools

VerifiedTrustedCommunity

# Lobster Lobster executes multi-step workflows with approval checkpoints. Use it when: - User wants a repeatable automation (triage, monitor, sync) - Actions need human approval before executing (send, post, delete) - Multiple tool calls should run as one deterministic operation ## When to use Lobster | User intent | Use Lobster? | | ------------------------------------------------------ | --------------------------

357,764SKILL.mdUpdated Apr 10, 2026

openclaw/extensions/lobster

Download

For Claude Desktop. Download once, then upload the file in the app — no terminal needed.

Need help? View full Cowork setup guide →

Install manually

Choose your platform

# Clone the repo
git clone https://github.com/lifemate-ai/familiar-ai.git

# Copy into Claude Code skills folder (global)
cp -r familiar-ai/dev/familiar-add-tool ~/.claude/skills/

Claude Code Skills — official skills path docs.

Repository

lifemate-ai/familiar-ai

75 stars

Compatible with

Claude Code

OpenAI Codex CLI

ChatGPT

Adoption

lifemate-ai/familiar-add-tool

$ install --global

Security Scan Results

SKILL.md

familiar-add-tool

When to Use

What This Skill Does

Step 1: Gather Requirements

Step 2: Create the Tool File

Step 3: Register in agent.py

3a. Import (top of file)

3b. _init_tools() — instantiate if config is available

3c. _all_tool_defs property — expose definitions

3d. _execute_tool() — route calls

Step 4: Update config.py

Step 5: Update System Prompt in agent.py

Step 6: Update .env.example

Checklist Before Finishing

Related Skills

lifemate-ai/familiar-debug-loop

lifemate-ai/familiar-check-env

openclaw/taskflow

openclaw/extensions/lobster

lifemate-ai/familiar-add-tool

$ install --global

Security Scan Results

SKILL.md

familiar-add-tool

When to Use

What This Skill Does

Step 1: Gather Requirements

Step 2: Create the Tool File

Step 3: Register in agent.py

3a. Import (top of file)

3b. _init_tools() — instantiate if config is available

3c. _all_tool_defs property — expose definitions

3d. _execute_tool() — route calls

Step 4: Update config.py

Step 5: Update System Prompt in agent.py

Step 6: Update .env.example

Checklist Before Finishing

Related Skills

lifemate-ai/familiar-debug-loop

lifemate-ai/familiar-check-env

openclaw/taskflow

openclaw/extensions/lobster

Step 3: Register in `agent.py`

3b. `_init_tools()` — instantiate if config is available

3c. `_all_tool_defs` property — expose definitions

3d. `_execute_tool()` — route calls

Step 4: Update `config.py`

Step 5: Update System Prompt in `agent.py`

Step 6: Update `.env.example`

Step 3: Register in `agent.py`

3b. `_init_tools()` — instantiate if config is available

3c. `_all_tool_defs` property — expose definitions

3d. `_execute_tool()` — route calls

Step 4: Update `config.py`

Step 5: Update System Prompt in `agent.py`

Step 6: Update `.env.example`