anvia-pipeline
Build deterministic multi-step work with Anvia Pipelines — steps, composition, parallel branches, decision routing, batch runs, agent, extractor, and decision…
Skill by anvia-hq
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/anvia-hq-anvia-pipeline
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: data, agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Shell scripts
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/anvia-pipeline
What's inside
- Skill name
anvia-pipeline- Description
- Build deterministic multi-step work with Anvia Pipelines — steps, composition, parallel branches, decision routing, batch runs, agent, extractor, and decision stages.
- License
- MIT
- Instructions
- 53 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
Files (4)
- SKILL.md
- agents-extract.md · references/
- steps-compose.md · references/
- check-pipeline.sh · scripts/
SKILL.md (first lines)
# Anvia Pipeline Skill
Use this skill when the work is a fixed sequence of steps, not a model-driven
loop: transforms, parsing, scoring, fan-out/fan-in, batch runs, or pipelines
with one bounded agent or extractor stage inside.
## Pipeline vs agent
- Reach for `Pipeline` when you can name the steps upfront. Reach for `Agent`
(see the `anvia-agent` skill) when the model must decide what to do next.
- The common failure is an agent with rising `maxTurns` doing a fixed sequence
— rewrite that as a pipeline with one `.agent()` stage instead.
If the stages must survive process restarts, use the `anvia-durable` skill to choose a
static agent DAG or versioned custom tasks. Calling a core Pipeline inside a task does
not checkpoint its individual stages.
## Process
1. Model the flow as stages (`references/steps-compose.md`) — step, compose,
parallel, batch.
2. Add model stages only where judgment is needed (`references/agents-extract.md`).
3. Run `scripts/check-pipeline.sh` from the app root before claiming done.
## Minimal slice
```ts
import { Pipeline } from "@anvia/core/pipeline";
import { z } from "zod";
const normalizeIncident = new Pipeline({ id: "normalize-incident", inputSchema: z.string() })
.step({ id: "trim", run: ({ input }) => input.trim() })
.step({ id: "collapse-whitespace", run: ({ input }) => input.replace(/\s+/g, " ") })
.step({
id: "summarize",
run: ({ input }) => ({
normalized: input,
wordCount: input.split(" ").length,
priority: input.toLowerCase().includes("outage") ? "high" : "normal",
}),
});
const result = await normalizeIncident.run({ input: " Checkout outage reported. " });
console.log(result.output);
```
## Output
Keep stages small, named, and typed — each stage's output is the next stage's
input. Point to the relevant reference file instead of pasting its contents
into chat.From anvia-hq/anvia, MIT.
More from anvia-hq/anvia
- 0.670score
anvia-durable
anvia-hq
Build recoverable Anvia workflows with @anvia/durable — durable main agents, owned subagents, task graphs, tool/effect recovery, persisted waits, and remote…
SkillAFreshness: A★ 57Claude Code +9 - 0.667score
anvia-channels
anvia-hq
Connect Discord, Slack, and Telegram to Anvia agents — channel adapters, agent-backed bots, proactive delivery, sessions, streaming, and platform actions.
SkillAFreshness: A★ 57Claude Code +9 - 0.667score
anvia-agent
anvia-hq
Build Anvia agents and tools — Agent options, tool definitions, approvals, memory, streaming, multi-agent teams, and provider wiring.
SkillAFreshness: A★ 57Claude Code +9 - 0.667score
anvia-evals
anvia-hq
Evaluate Anvia agents and retrieval — deterministic metrics, semantic similarity, LLM judges, RAG quality, and CLI eval runs.
SkillAFreshness: A★ 57Claude Code +9 - 0.667score
anvia-studio
anvia-hq
Run and inspect Anvia agents in Studio — local serving, playground, traces, approvals, sessions, evals, and observability.
SkillAFreshness: A★ 57Claude Code +9 - 0.664score
anvia-mcp
anvia-hq
Connect MCP servers to Anvia agents — clients, transports, tool discovery, URL safety, and lifecycle.
SkillAFreshness: A★ 57Claude Code +9 - 0.664score
anvia-rag
anvia-hq
Build retrieval and RAG with Anvia — chunking, embeddings, vector stores, knowledge graphs, filters, and search tools.
SkillAFreshness: A★ 57Claude Code +9 - 0.664score
anvia-chat
anvia-hq
Wire an Anvia chat UI end to end — server stream route, client transport, React hook, and UI primitives.
SkillAFreshness: A★ 57Claude Code +9 - 0.659score
release-notes
anvia-hq
Draft concise product release notes from a list of shipped changes.
SkillAFreshness: A★ 57Claude Code +9
Signals
- Stars
- 57
- Forks
- 6
- Open issues
- 0
- Last commit (this folder)
- Oct 16, 2026 (12 days ago)
- Last release
- Oct 23, 2026 (5 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/anvia-hq/anvia
- Website
- anvia.dev
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from anvia-hq
- 0.670score
anvia-durable
anvia-hq
Build recoverable Anvia workflows with @anvia/durable — durable main agents, owned subagents, task graphs, tool/effect recovery, persisted waits, and remote…
SkillAFreshness: A★ 57Claude Code +9 - 0.667score
anvia-agent
anvia-hq
Build Anvia agents and tools — Agent options, tool definitions, approvals, memory, streaming, multi-agent teams, and provider wiring.
SkillAFreshness: A★ 57Claude Code +9 - 0.667score
anvia-channels
anvia-hq
Connect Discord, Slack, and Telegram to Anvia agents — channel adapters, agent-backed bots, proactive delivery, sessions, streaming, and platform actions.
SkillAFreshness: A★ 57Claude Code +9 - 0.667score
anvia-evals
anvia-hq
Evaluate Anvia agents and retrieval — deterministic metrics, semantic similarity, LLM judges, RAG quality, and CLI eval runs.
SkillAFreshness: A★ 57Claude Code +9 - 0.667score
anvia-studio
anvia-hq
Run and inspect Anvia agents in Studio — local serving, playground, traces, approvals, sessions, evals, and observability.
SkillAFreshness: A★ 57Claude Code +9 - 0.664score
anvia-chat
anvia-hq
Wire an Anvia chat UI end to end — server stream route, client transport, React hook, and UI primitives.
SkillAFreshness: A★ 57Claude Code +9
Cite this page
anvia-pipeline skill. Agent Mart. https://agentmart.dev/skills/anvia-hq-anvia-pipeline (last updated 2026-10-24)