session-hub
Skill39 starsChanged 55 days agoMITView on GitHub
Search and reuse past coding-agent sessions across harnesses (Pi, Claude Code, Codex, OpenCode, Crush, JCode). Use when the user asks which session they worked on something in, wants to find prior context, wants to continue work from another tool, or asks "where did I do X" / "in what chat did I fix Y".
6 other projects use a version of this.
Adopted by
Most recently taken up by
- ryanportfolio/winuse-mcptheir own version, first seen Nov 8, 2026
- ryanportfolio/Harness-Firmwaretheir own version, first seen Nov 7, 2026
- ryanportfolio/HarnessFirmware.comtheir own version, first seen Nov 5, 2026
- ryanportfolio/Corewise.Academytheir own version, first seen Nov 5, 2026
- ryanportfolio/STKtheir own version, first seen Oct 20, 2026
- ryanportfolio/savetokenstheir own version, first seen Oct 20, 2026
Used with
Projects with it also use addskill, apply-firmware and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
ryanportfolioSkilladdskill
Create, import, update, or install repo or personal skills: runtime ownership, resources, discovery checks.
Also in 6 projects that use this
- ryanportfolioSkill
apply-firmware
Bring current folder (plain folder, existing repo, old or partial firmware copy) up to full Harness Firmware in place. Mirror external repo → adopt-repo.
Also in 6 projects that use this
- +10 more from ryanportfolio
Other versions6 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 6
- ryanportfolio/winuse-mcp★ 2same name
- ryanportfolio/Corewise.Academy★ 1same name
- ryanportfolio/Harness-Firmware★ 1same name
- ryanportfolio/savetokens★ 1same name
- ryanportfolio/STK★ 1same name
- ryanportfolio/HarnessFirmware.com★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 1 is distinctive, and 1 is very specific to this project.
0
Common
Most of the 6 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 20
Do not use it for reading project source files.
- Very specific: line 61
Nothing outside
~/.pi/agent/pi-session-hub/is written: Do not attempt to modify any harness's session store.
“Common” is measured: kept by at least 6 in 10 of the 6 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Session Hub
pi-session-hub indexes the local session stores of every coding harness on this
machine and exposes them through one list.
When to use it
Reach for the hub when the user:
- asks which session they worked on something in ("where did I implement the db layer?")
- wants prior context before starting new work
- wants to continue a conversation that started in another tool
- asks what they were doing on a project recently
Do not use it for reading project source files. It reads agent transcripts, not code.
Tools and commands
| Surface | Use |
session_hub_search tool | Search the local index. Preferred: you can call it directly. |
/session-hub | Open the interactive two-pane browser (also /hub, or alt+r). |
/session-search <query> | Same search, printed into the chat. Supports --harness <id> and --limit <n>. |
/session-open <id> | Read-only metadata + transcript preview. |
/session-handoff <id> | Handoff draft in the editor, to review before sending. |
/session-native <id> | Resume the session in its original harness. Asks for confirmation first. |
In the hub, Enter loads the selected session's conversation into the current chat
as a collapsed imported transcript message. The context is tiered and budgeted
(~10k tokens by default): recent turns verbatim, older ones condensed to one line,
tool output compressed, and anything omitted is counted rather than hidden. v
reads the transcript for free, o switches Pi to a Pi session, n reopens it in
its original tool. The hub is a full-screen replacement UI; ? shows the keyboard
reference.
Harness ids: pi, claude-code, codex, opencode, crush, jcode.
Search query syntax
- bare terms use prefix matching:
pliegomatchespliego-prod "exact phrase"for phrases-termto exclude
Working rules
- Never claim a session is Pi-native when it is not. Session ids are namespaced
harness:nativeId. Report the harness honestly. - Prefer loading the context over pasting transcript text.
Enterin the hub (orsession_hub_context) produces a structured document that is labelled as imported and lists what the source could not provide. - Respect "not available". Some sources do not record changed files, commands or working directories. Report that as unavailable rather than guessing.
/session-nativefor OpenCode is intentionally unavailable. There is no verified resume-by-id command, so the hub refuses rather than inventing one.- Nothing outside
~/.pi/agent/pi-session-hub/is written. Do not attempt to modify any harness's session store.
Rebuilding the index
The index is rebuilt automatically when empty. To force a refresh, open /session-hub
and press r. The index lives at ~/.pi/agent/pi-session-hub/index.sqlite.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.