plexus 0.2.0 — graph export, runnable pipelines, grounded comparison#1
Merged
Conversation
- graph.py: to_mermaid / to_dot + `plexus graph` to see the whole mesh. - run.py: pipeline_script + `plexus run --goal` turns a plan into a runnable, ordered shell script ending at the target; feedback loops emitted as a comment. - COMPARISON.md: honest positioning vs MCP / LangGraph / Dagster / CrewAI. Wins the decentralized, evidence-cited discovery niche; does not execute (by design), composing above an executor rather than replacing it. - 22 falsifiers (was 17). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
…he format) - export_all / `plexus export` writes each flagship's <organ>.interop.json. - The five real manifests committed under manifests/ — the exact files a tool ships to join the mesh. - Round-trip test: discovery from the JSON files == the in-code registry, so the public format losslessly represents a real tool (the premise of decentralized discovery). 24 falsifiers. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
`plexus mcp` serves discover/wiring/plan/route as MCP tools over stdio JSON-RPC (the shape every flagship speaks). A harness can now ask "what feeds crucible?" or "how do gather and crucible connect?" while it works, instead of only from a human CLI. handle() is transport-free and fully tested; 31 falsifiers. This is the "make each tool aware of one another, so they immediately plugin" step: the interop mesh is now itself a pluggable tool. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Extends plexus from discovery-only to a demoable, actionable, honestly-positioned tool.
What's new
graph.py,plexus graph --format mermaid|dot) — render the whole mesh as a diagram; self-loops marked distinctly.run.py,plexus run --goal ORGAN) — turn a plan into a runnable, ordered shell script that ends at the target. Plans become action, not advice. Feedback loops are emitted as a comment, never linearized into a false order.COMPARISON.md— grounded positioning vs MCP / LangGraph / Dagster / CrewAI. Honest: plexus wins the decentralized, evidence-cited discovery niche and does not execute (by design), composing above an executor rather than replacing it.Tests
22 falsifiers (was 17), all green locally. CI runs the 3-OS × 3-Python matrix + wheel install.
🤖 Generated with Claude Code