Preserve result.trace_id across plugin HTTP wrappers, include it in commit success and failure logs, and surface it in user-visible commit confirmations where supported.
* feat(plugins): align opencode and pi memory integrations
* fix(installer): tolerate missing optional harness CLIs
* fix(installer): install opencode file wrapper
* fix(opencode): import path for logger initialization
* fix(installer): register pi extension after copy
* feat(plugins): use MCP for opencode integration
* docs: move OpenCode and pi integrations to dedicated pages
Promote the OpenCode plugin and pi extension out of the community-plugins
page into their own numbered agent-integrations pages (10-opencode, 11-pi,
en + zh), update the overview routing table, and refresh the OpenCode image
cards to the hybrid MCP architecture (unified installer, openviking_* MCP
tools, ovcli.conf credentials).
* docs: bare TOS installer commands and reference more examples
Drop --harness from TOS-mirror install commands (image cards use the bare
installer URL, matching the claude-code/codex cards); add Open WebUI tool
server and an examples/ pointer to the community-plugins page (en + zh).
* docs: bare TOS installer commands across agent-integration pages
TOS-mirror install commands carry no flags anywhere; the installer wizard
asks for source, harnesses, language, and credentials.
* feat(examples): add pi coding agent memory extension
Add OpenViking extension for the pi coding agent (github.com/mariozechner/pi-coding-agent).
Provides synchronous recall, auto-capture, 7 LLM tools, session commit, and memory indexing.
Architecture: 8 TypeScript files + config, loaded by pi's jiti transpiler — zero dependencies
beyond Node.js. Design informed by all three existing OV plugins (OpenClaw for synchronous recall,
Claude Code for production-hardened capture/ranking, Hermes for anti-patterns).
Key features:
- Synchronous recall via pi's context event (not stale prefetch)
- 7 native LLM tools (search, read, browse, remember, forget, add_resource, archive_expand)
- Tool USE preservation, tool result dropping
- Token-budgeted content resolution with graceful degradation
- Memory pollution prevention (context block stripping)
- Client-driven commit threshold
- Pre-compact commit, session resume rehydration
- Memory index (map model vs flashlight)
- Write queue with batching
- Globs-based bypass patterns
- Env var config overrides (OPENVIKING_URL, OPENVIKING_API_KEY, etc.)
~1692 lines TypeScript + 29 lines config. Tested with local OpenViking server:
7/7 tests passing (load, sync, commit, recall, 7 tools, /viking command, continuation).
* docs(examples): add DESIGN.md with cross-plugin comparison and rationale
Adds the full design specification (901 lines) documenting:
- Cross-plugin comparison table (Hermes vs OpenClaw vs Claude Code vs Pi)
- Architectural decisions and their provenance
- Detailed event flow diagrams
- Capture filtering, memory stripping, token estimation
- Write queue design and async patterns
- Why a memory index (map vs flashlight model)
Useful as a reference for building OV extensions for any agent harness,
not just pi.
t0saki