Run CIAgent regression checks after changing an AI agent's code, prompts, or knowledge base in a repo that has agentci_spec.yaml, and interpret the results. Use after editing agent logic, before committing agent changes, or when the user asks whether the agent still works.
インストールする前に、エージェントに与えられる指示の中身を確認できます。
The repo has agentci_spec.yaml (if it does not, use the onboard skill
instead). Your job: run the right check for the change that was just made,
read the result correctly, and never paper over a failure.
| Situation | Command |
|---|---|
| Spec or wiring changed, or no API keys | ciagent test --mock |
| Agent code / prompt / retrieval changed | ciagent test --yes --format json |
| Result differs from last run, or flakiness suspected | ciagent test --runs 3 --yes |
| Knowledge base changed | ciagent generate-checks --dry-run, review, then apply |
| The LLM judge's verdicts look wrong | ciagent judge-audit |
Live runs (test without --mock, judge-audit, generate-checks) call model
APIs on the user's keys. Mock mode is free. If the user has not already
approved live runs in this session, prefer --mock or ask.
Exit codes: 0 pass (including flaky-but-passing), 1 correctness failure
(with --runs N: failed in every run), 2 infra or config error — fix the
setup, not the agent.
With --format json: per-query entries carry layer results (correctness /
path / cost) and the answer text; with --runs N a top-level stability block
lists flipped queries with flip_source.
Flip sources route the work:
agent-variance — the agent's answer changed between runs → fix the agent
(prompt, retrieval, temperature).judge-flake — same answer, the LLM judge changed its verdict → fix the
eval (tighten the rubric or replace with a deterministic check).infra-error — a judge API call failed → retry; fix nothing.mixed — ambiguous; look at the answers yourself.ciagent bootstrap --runner <runner> --queries <file> --yes for that query,
or update the spec's expectations — with the user's confirmation.まだレビューはありません。使ってみた感想をお寄せください。
概要と使いどころ
Automate ActiveCampaign tasks via Rube MCP (Composio): manage contacts, tags, list subscriptions, automation enrollment, and tasks. Always search tools first for current schemas.
日本語の概要は準備中です。原文の説明を表示しています。
Analytics your AI agent can actually use. Track, analyze, run A/B experiments, and optimize across all your projects via CLI. Includes a growth playbook so your agent knows HOW to grow, not just what to track.
日本語の概要は準備中です。原文の説明を表示しています。
Teaches when to recall from long-term memory before acting and when to save durable decisions, corrections and failures afterwards. Use when a memory tool or MCP memory server is connected but the agent is not using it consistently, when the user complains that the assistant forgets preferences, conventions or past decisions between sessions, or when setting up persistent memory for a project. Works with any memory backend: a folder of Markdown files, a local MCP server, or a managed service.
日本語の概要は準備中です。原文の説明を表示しています。
Audit local AI coding-agent sessions with agenttrace for cost, tokens, tool failures, latency, anomalies, health, diffs, and CI gates.
日本語の概要は準備中です。原文の説明を表示しています。
Audit whether a website can be found, crawled, and cited by AI answer engines such as ChatGPT Search, Perplexity, Google AI Overviews, and Microsoft Copilot. Use when someone asks why their brand is missing from AI answers, whether AI crawlers can read their site, how to get cited by ChatGPT or Perplexity, or asks for a GEO or AEO (generative / answer engine optimization) review. Produces a citation baseline across buyer-intent prompts, a crawler-access check, a citability review of named pages, and a ranked fix list. Not for keyword rank tracking, paid search, or pages behind a login.
日本語の概要は準備中です。原文の説明を表示しています。
Automate Airtable tasks via Rube MCP (Composio): records, bases, tables, fields, views. Always search tools first for current schemas.
日本語の概要は準備中です。原文の説明を表示しています。