Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/shyftlabs/continuum/continuum-agentnpx skills add shyftlabs/continuum --skill continuum-agentgit clone --depth 1 https://github.com/shyftlabs/continuumWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/shyftlabs/continuum/continuum-agent)<a href="https://agentmods.dev/skills/shyftlabs/continuum/continuum-agent"><img src="https://agentmods.dev/badge/skills/shyftlabs/continuum/continuum-agent.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00074 | $0.01287 |
| Opus 5 | $0.00037 | $0.00643 |
| Sonnet 5 | $0.00015 | $0.00257 |
| Haiku 4.5 | $0.00007 | $0.00129 |
Grade A, and why
continuum-agent scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 152 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Continuum Agent Skill
Authoritative source: docs/agent.md.
Core API
from continuum.agent import BaseAgent, AgentRunner
from continuum.agent.config import AgentConfig, AgentMemoryConfig, RunnerConfig
from continuum.agent.types import (
Handoff, MemoryScope, RunContext, AgentResponse, EventType,
)
Minimal agent
agent = BaseAgent(
name="my-agent", # required, [A-Za-z0-9_-]+
instructions="...", # supports {slot} templates
model="gpt-4o-mini", # provider chosen by prefix
temperature=0.7,
max_tokens=None,
)
runner = AgentRunner()
resp = await runner.run(agent, "user input", user_id="u1", session_id="s1")
print(resp.content)
print(resp.usage.total_tokens)
print(resp.structured_output) # populated if output_schema set
Important BaseAgent parameters
| Parameter | Default | Notes |
|---|---|---|
name |
required | unique id |
instructions |
"" |
supports {template_vars} |
model |
settings.default_llm_model |
prefix routes provider |
tools |
[] |
LLM-shaped dicts |
mcp_servers |
[] |
Auto-discovered tool sources |
handoffs |
[] |
[Handoff(target_agent=..., description=...)] |
memory_config |
AgentMemoryConfig() |
search_memories=True / store_memories=True |
output_schema |
None |
Pydantic model |
template_vars |
{} |
static slots |
examples |
[] |
few-shot |
instruction_modifiers |
[] |
dynamic prompt rewriters |
on_start / on_end / on_error / on_tool_call / on_handoff |
None |
sync callables |
config |
AgentConfig() |
max_turns=25, react_mode, reasoning_mode, scanners, strict_security, max_consecutive_handoffs=3 (same request to the same target before HandoffLoopError; fan-out over N items is never blocked), … |
gateway_mode |
None |
Smart Gateway routing: "strict" / "modest" / "quality" |
policy_store |
None |
Access-control store. None = fail-open (no authorization). Use PolicyStore.default_deny([...]) to fail closed; set AgentConfig(strict_security=True) to raise when side-effectful tools have no policy |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- today First seen · 152 lines · 74 tokens per session scan A 3914b6816f73
continuum-agent is a skill published in the GitHub repository shyftlabs/continuum (84 stars, last pushed today), licensed Apache-2.0. It adds 74 tokens to every session and 1,287 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
reachai-onboarding
Integrate Java business systems with ReachAI SDK registration, SDK instance heartbeat, gateway/embed access, and optional API Management handoff. Use when asked to connect a Spring Boot service to ReachAI, add reachai-capability-sdk or reachai-spring-boot2-starter, configure…
agenticx-agent-builder
Guide for creating persistent Near desktop digital avatars (数字分身) via natural-language interview and the createavatar tool. Use when the user wants to create an avatar, digital twin, specialist agent persona, or add someone to the avatar sidebar.
agenticx-skill-manager
Guide for managing AgenticX skills including listing, searching, installing, uninstalling, publishing, and running a skill registry server. Use when the user wants to manage skills, find available skills, publish custom skills, set up a skill registry, or understand the skill ecosystem.
agenticx-tool-creator
Guide for creating custom tools in AgenticX including function decorator tools, MCP tool integration, tool registries, and remote tool access. Use when the user wants to create tools for agents, integrate external APIs as tools, build MCP servers, or extend agent capabilities with custom functions.
skill-creator
Create or update AgenticX skills from a live conversation workflow. Use when the user asks to save, persist, or encapsulate a multi-step procedure as a local skill, when tool calls were repeated too often, or when refining SKILL.md frontmatter and discoverability after skillmanage.
feature-loop
Coding agent for the project-level harness — one feature per session, must verify, must be mergeable.