Official Flowlines plugins for coding agents. One repository serves as a plugin marketplace for both Claude Code and Codex CLI, and ships the flowlines plugin:
| Component | What it does |
|---|---|
flowlines MCP server |
Connects your agent to your Flowlines workspace at https://api.flowlines.ai/mcp. Ask what your agents' users did, what changed since a release, where sessions go wrong, and record findings as notes. |
flowlines-mcp-observability skill |
Instruments an MCP server so its tool calls arrive in Flowlines as canonical MCP telemetry, through AGNTCY Observe or vanilla OpenTelemetry. |
flowlines-agent-observability skill |
Installs, repairs, diagnoses, or removes user-level Flowlines telemetry for Claude Code and Codex CLI sessions on macOS and Linux. |
flowlines-weekly-review skill |
A periodic review over the MCP server: what changed since the last review, signals, outcome movements, and what to pin for next time. |
flowlines-release-check skill |
Before-and-after comparison of an agent release: outcomes, intents, cost, signals, and evidence sessions, with the right denominators. |
flowlines-investigate-session skill |
From a signal, a user, or a complaint to the failing turn, with minimal exposure of end-user content. |
flowlines-cohort-builder skill |
Sizes a user segment, writes it in the Flowlines cohort rule vocabulary, and compares it against a baseline. |
flowlines-doctor skill |
Diagnoses missing or incomplete data across every source: coding-agent telemetry, instrumented MCP servers, LangSmith and Langfuse connectors, and OTLP applications. |
- The MCP server reads production conversations between end users and your agents. Treat everything it returns as confidential; it never writes to your namespace except through the explicit
save_noteandreport_outcometools. flowlines-mcp-observabilityexports validated tool arguments, client-visible results, and user identity metadata from the instrumented server to Flowlines. That data can contain personal data, customer data, source code, or other sensitive content.flowlines-agent-observabilityexports full prompts, assistant messages, tool inputs, and tool outputs from your machine to Flowlines.
Both skills ask for explicit consent before changing anything, and neither prints or stores your Flowlines API key in chat. They need a namespace API key, created in the Flowlines app under Settings, API keys; the skills point you there and can open the page for you.
claude plugin marketplace add flowlines-ai/plugins && claude plugin install flowlines@flowlinesThen run /mcp inside Claude Code and sign in to flowlines. Skills are available as /flowlines:<skill-name>, for example /flowlines:flowlines-weekly-review or /flowlines:flowlines-doctor. Add --scope project to the install command to enable the plugin for one repository only.
codex plugin marketplace add flowlines-ai/plugins && codex plugin add flowlines@flowlinesThen sign in with codex mcp login flowlines, or open /plugins inside Codex. Skills are available as $<skill-name>, for example $flowlines-weekly-review or $flowlines-doctor.
The plugin registers an MCP server named flowlines. If you previously added the server by hand under the same name, remove that entry to avoid a duplicate.
Claude Code reads marketplaces and plugins from a repository's .claude/settings.json:
{
"extraKnownMarketplaces": {
"flowlines": { "source": { "source": "github", "repo": "flowlines-ai/plugins" } }
},
"enabledPlugins": { "flowlines@flowlines": true }
}Codex reads a repository-level marketplace from .agents/plugins/marketplace.json:
{
"name": "my-team",
"interface": { "displayName": "My team" },
"plugins": [
{
"name": "flowlines",
"source": {
"source": "git-subdir",
"url": "https://github.com/flowlines-ai/plugins.git",
"path": "plugins/flowlines"
},
"policy": { "installation": "INSTALLED_BY_DEFAULT", "authentication": "ON_USE" }
}
]
}.claude-plugin/marketplace.json Claude Code marketplace
.agents/plugins/marketplace.json Codex marketplace
plugins/flowlines/
.claude-plugin/plugin.json Claude Code manifest
.codex-plugin/plugin.json Codex manifest and directory listing
.mcp.json MCP server shared by both clients
skills/ Skills shared by both clients
Both clients read the same .mcp.json and skills/ tree. Each client keeps its own manifest because Claude Code only reads .claude-plugin/plugin.json and Codex adds directory metadata under interface.
Validate the manifests with the real CLIs, then the skill packages and the installer:
scripts/validate_plugins.sh
python3 scripts/validate_skills.py
plugins/flowlines/skills/flowlines-agent-observability/scripts/test_installer.shvalidate_plugins.sh needs claude and codex on your PATH. It runs offline: Claude validates the manifests and Codex installs the plugin from this checkout into a throwaway CODEX_HOME.
To try the plugin from a checkout without installing it, run claude --plugin-dir plugins/flowlines, or add this directory as a local marketplace with codex plugin marketplace add ..
- Bump
versioninplugins/flowlines/.claude-plugin/plugin.json,plugins/flowlines/.codex-plugin/plugin.json, and the plugin entry in.claude-plugin/marketplace.json. - Merge to
main. Marketplace installs trackmain; users pick up the new version withclaude plugin update flowlines@flowlinesorcodex plugin marketplace upgrade. - Tag the release with
claude plugin tag plugins/flowlines.
The skills were moved here from flowlines-ai/mcp-server-observability and flowlines-ai/coding-assistant-observability.
Questions or issues: support@flowlines.ai, or open an issue in this repository.
MIT