@mariozechner/pi-coding-agent
0.73.1 · 7 May 2026MIT443.1k / weekRepositoryHomepagenpmAtom feed
0.73.1 is the newest release of 0.56.x.
In short
6 new9 fixed
Security
3 advisories affect 0.56.3
- HighPi Agent: Predictable temporary extension install paths allow local privilege escalation on shared Linux hostsGHSA-jfgx-wxx8-mp94 · 17 Jun 2026No fix yet
- LowPi Agent: Race condition in Pi auth.json writes could expose stored credentialsGHSA-r95r-rj6r-c39x · 17 Jun 2026No fix yet
- LowPi Agent: Potential XSS in HTML session exports via Markdown URL sanitization bypassGHSA-7v5m-pr3q-6453 · 16 Jun 2026No fix yet
Release note
New Features
claude-sonnet-4-6model available via thegoogle-antigravityproviderCustom editors can now define their own
onEscape/onCtrlDhandlers without being overwritten by app defaults, enabling vim-mode extensionsShift+Enter and Ctrl+Enter now work inside tmux via xterm modifyOtherKeys fallback (docs/tmux.md, #1872)
Auto-compaction is now resilient to persistent API errors (e.g. 529 overloaded) and no longer retriggers spuriously after compaction (#1834, #1860)
Added
Added
claude-sonnet-4-6model for thegoogle-antigravityprovider (#1859).Added tmux setup documentation for modified enter key support
Fixed
Fixed custom editors having their
onEscape/onCtrlDhandlers unconditionally overwritten by app-level defaults, making vim-style escape handling impossibleFixed auto-compaction retriggering on the first prompt after compaction due to stale pre-compaction assistant usage
Fixed sessions never auto-compacting when hitting persistent API errors (e.g. 529 overloaded) by estimating context size from the last successful response
Fixed compaction summarization requests exceeding context limits by truncating tool results to 2k chars
Fixed
/newleaving startup header content, including the changelog, visible after starting a fresh sessionFixed misleading docs and example implying that returning
{ isError: true }from a tool'sexecutefunction marks the execution as failed; errors must be signaled by throwingFixed model switches through non-reasoning models to preserve the saved default thinking level instead of persisting a capability-forced
offclampFixed parallel pi processes failing with false "No API key found" errors due to immediate lockfile contention on
auth.jsonandsettings.jsonFixed OpenAI Responses reasoning replay regression that broke multi-turn reasoning continuity
A whole project to check? Paste its package.json in the dependency check: every deprecated, outdated or vulnerable dependency, with what to upgrade to. Free, no account.