// Refresh the list of agent frameworks supported by Arize tracing and diff against what's already in the repo. Pulls live data from https://arize.com/docs/llms.txt and produces a clean to-do list. Part of the rosetta-add-framework flow; can also be invoked standalone to answer "what frameworks are left to add?"
Add a new agent framework to the Rosetta Stone repo — researches the framework, builds all three observability tiers (no-observability, phoenix, ax), tests each, runs a Playwright smoke against the UI, updates README + TODO, and raises a PR. Trigger when the user asks to "add the <framework> framework", "implement <framework>", "wire up <framework>", or similar. The framework must be one of the Arize-supported agent frameworks (see TODO below).
Build a single tier (no-observability, phoenix, or ax) for a new framework. Clones the closest existing tier, swaps in framework-specific agent.py / tools.py / requirements.txt, and (for observability tiers) adds tracing.py + main.py wiring + eval-harness scripts. Part of the rosetta-add-framework flow; can be invoked standalone to rebuild a single tier from scratch.
Run a public-flow Playwright smoke test against a freshly-built framework tier's Next.js frontend. Covers home page rendering + product browsing — the parts that don't require X/Twitter OAuth. The Playwright project (package.json, config, tests) lives inside this skill directory and is checked into the repo. Part of the rosetta-add-framework flow.
Finalise a newly-added framework — updates the README's supported-frameworks table, directory tree, and per-framework "what differs" section, marks off the framework in the orchestrator skill's embedded TODO, commits per tier, and raises a PR. Part of the rosetta-add-framework flow.
Test a freshly-built tier for a new framework — boots the backend, smoke-tests the chat endpoint, runs synthetic requests, and (for phoenix) runs the eval harness. Verifies traces land in the right project. Part of the rosetta-add-framework flow; can be invoked standalone after a build to validate.
Record a full Wonder Toys demo by running a canned 3-turn conversation (search dragons → buy plushie → ship), then opening Arize AX in Safari and screenshotting the session view plus every trace in it. Use when the user asks to "capture a demo", "record screenshots of an Arize session", "demo the agent flow", or any similar phrasing. macOS only — uses AppleScript and `screencapture`.
| name | rosetta-add-framework-discover |
| description | Refresh the list of agent frameworks supported by Arize tracing and diff against what's already in the repo. Pulls live data from https://arize.com/docs/llms.txt and produces a clean to-do list. Part of the rosetta-add-framework flow; can also be invoked standalone to answer "what frameworks are left to add?" |
Fetches the live list of agent-framework integrations from Arize's docs index, diffs against the repo, and prints what's left.
# Use WebFetch on https://arize.com/docs/llms.txt with this prompt:
# "From this index, list every page whose path includes
# 'python-agent-frameworks' or 'ts-js-agent-frameworks'. Group by
# language. Output one slug per line."
The slug is the directory segment in the URL (e.g. crewai, langchain, pydantic, google-adk).
cd /Users/jimbobbennett/github/project-rosetta-stone
# Existing framework directories under ax/ (canonical — all 3 tiers stay in sync)
ls -d ax/*-py/ ax/*-js/ ax/*-ts/ ax/mastra/ ax/vercel-ai-sdk/ 2>/dev/null \
| xargs -n1 basename \
| sed -E 's/-py$|-js$|-ts$//' \
> /tmp/rosetta-existing
# Compare against the slug list from step 1
Map each Arize slug to one of:
[x]) — in /tmp/rosetta-existing[~]) — viability gate decided not to build (e.g. voice-only, no Anthropic). The reason is appended after the framework name in the orchestrator's TODO. Preserve these markers — don't reset them to [ ].[!]) — last build attempt broke. The orchestrator may retry on a future run; preserve the marker until the failure is resolved.[ ]) — recommended for the repo (CrewAI, Pydantic AI, LangChain, LlamaIndex, etc.)When refreshing the TODO snapshot (step 5 below), preserve existing [x], [~], and [!] markers verbatim — including the reason text after them. Only flip newly-implemented frameworks from [ ] to [x].
Print a checklist:
Frameworks left to add:
Python (Tier A):
[ ] agno https://arize.com/docs/ax/integrations/python-agent-frameworks/agno/agno-tracing
[ ] autogen https://arize.com/docs/ax/integrations/python-agent-frameworks/autogen/autogen-tracing
[x] crewai (in repo as crewai-py)
...
TypeScript (Tier A):
[ ] beeai https://arize.com/docs/ax/integrations/ts-js-agent-frameworks/beeai/beeai-tracing-js
...
Borderline (decide before building):
[ ] guardrails-ai (output validation)
...
If the orchestrator's TODO snapshot is more than a week stale OR if the diff surfaces new frameworks Arize added, edit .claude/skills/rosetta-add-framework/SKILL.md and refresh the "## TODO" section + the "last refreshed" date.
If called from the orchestrator with a target framework, also emit:
TARGET: <framework>
TARGET_STATUS: not-implemented | already-implemented | not-supported-by-arize | borderline
TARGET_DOCS: <full URL to the Arize tracing doc>
If status is anything other than not-implemented, the orchestrator should abort.