@lebronj/pi-suite
JHP's Pi extension suite for team coding workflows
Package details
Install @lebronj/pi-suite from npm and Pi will load the resources declared by the package manifest.
$ pi install npm:@lebronj/pi-suite- Package
@lebronj/pi-suite- Version
0.1.12- Published
- Jun 16, 2026
- Downloads
- 1,262/mo · 1,262/wk
- Author
- lebronj
- License
- MIT
- Types
- extension, skill, prompt
- Size
- 383.2 KB
- Dependencies
- 3 dependencies · 5 peers
Pi manifest JSON
{
"extensions": [
"./extensions",
"./vendor/pi-memory/index.ts",
"node_modules/pi-mcp-adapter/index.ts",
"node_modules/pi-subagents/src/extension/index.ts",
"node_modules/pi-web-access/index.ts"
],
"prompts": [
"./prompts",
"node_modules/pi-subagents/prompts"
],
"skills": [
"./skills",
"node_modules/pi-subagents/skills",
"node_modules/pi-web-access/skills"
]
}Security note
Pi packages can execute code and influence agent behavior. Review the source before installing third-party packages.
README
@lebronj/pi-suite
JHP's Pi extension suite for team coding workflows.
Install
npm install -g --ignore-scripts @earendil-works/pi-coding-agent
pi install npm:@lebronj/pi-suite
Or use the bootstrap script to install Pi, configure the team OpenAI-compatible endpoint, install this suite, and set up Bun + qmd for memory search:
curl -fsSL https://registry.npmjs.org/@lebronj/pi-suite/-/pi-suite-0.1.12.tgz | tar -xzO package/scripts/bootstrap.sh | bash
What Is Included
- Local extensions: autogoal, goal mode, pet, prompt URL widget, snake, TPS notifications.
- Prompts: changelog audit, issue analysis, PR review, review workflow, commit workflow, wrap workflow.
- Skills: provider checklist, Pi capability reference, image-to-editable-PPT workflow.
- Vendored package:
@jhp/pi-memory, including qmd search, external curator service, and memory/skill-draft versioning.
pi-mcp-adapter, pi-subagents, and pi-web-access are loaded by this suite. Do not install them separately unless you filter suite resources, because duplicate extension tools/flags can conflict.
Figma is not installed or loaded by default. Enable it only when needed:
pi install npm:pi-mono-figma
# then run /reload or restart pi
Disable Figma later with:
pi remove npm:pi-mono-figma
# then run /reload or restart pi
Debug-only extensions are intentionally excluded:
dump-system-prompt.tszz-full-session-log.tsagentmemory
Autogoal
/autogoal <task> starts a bounded autonomous coding run. It persists the objective, auto-continues with loop budgets, checkpoints at high context usage, and can continue in a fresh session when the context window gets tight.
Useful commands:
/autogoal <task>
/autogoal status
/autogoal pause
/autogoal resume
/autogoal checkpoint optional reason
/autogoal drop
Behavior:
- 60% context: prepare and stay concise.
- 75% context: write a structured checkpoint soon.
- 85% context: checkpoint and switch to a new session.
- Completion requires current-state evidence: changed files read after edits and a passing validation command.
- Subagents are optional and budgeted; worker subagents must use worktree isolation.
- Run artifacts are written under
~/.pi/agent/workflow-runs/autogoal-<run-id>/.
Team Model Setup
The bootstrap script can be run with curl | bash: it reads the API key from the terminal instead of stdin. It asks for an API key and writes:
- Provider: OpenAI-compatible
openai - Base URL:
https://claude-code.club/openai/v1 - Default model:
gpt-5.5
The API key is written to ~/.pi/agent/models.json on the user's machine. Do not publish a shared key in this package.
Memory And Versioning
@jhp/pi-memory works without qmd for core memory features:
memory_writememory_readmemory_editscratchpadmemory_curate
memory_search needs qmd. The bootstrap script installs and initializes qmd when Bun is available. If Bun is missing, install qmd later:
bun install -g https://github.com/tobi/qmd
qmd collection add ~/.pi/agent/memory --name pi-memory
qmd embed
Memory versioning is enabled by default. It snapshots ~/.pi/agent/memory and ~/.pi/agent/skill-drafts into ~/.pi/agent/evolution, commits local changes automatically, and leaves push manual by default. memory_curate also scans yesterday's daily log into REVIEW.md when learning is enabled and the daily file changed since the last scan.
Useful commands:
/memory-version-status
/memory-version-snapshot optional reason
/memory-version-list
/memory-version-restore <snapshot-id> [memory|skill-drafts|all]
/memory-version-push
Memory evolution is local-only by default and does not configure a shared remote. If a user wants backup sync, set PI_EVOLUTION_REMOTE to their own private repo before bootstrap/setup, or add a personal remote later with git -C ~/.pi/agent/evolution remote add origin <url>. Set PI_EVOLUTION_AUTO_PUSH=1 only if automatic remote sync is desired.
Review And Commit Workflows
Use /review [target] [--fix] [--comment] [--summary] to inspect current diffs, branch ranges, or PRs with findings first and a verdict last. It is local-only by default: GitHub comments are drafted first and posted only after explicit confirmation.
Use /commit [message|split|pr|apply] to inspect current changes, warn about mixed concerns, draft commit messages, suggest split points, or draft PR text. It does not stage, commit, or push unless you explicitly confirm the exact action.
These workflows are prompt-template workflows only. They do not merge read behavior, add tool discovery, rewrite memory/skills, or run as hidden background processes.
Goal Mode
Use /goal <objective> to keep Pi working on one task until it is verified complete. Goal mode injects hidden task context, enables a goal tool for completion/drop/resume, and auto-continues between turns instead of stopping at a minimal implementation.
Useful commands:
/goal <objective>
/goal show
/goal pause
/goal resume
/goal drop
/goal auto on
/goal auto off
Update
pi update
Only update Pi itself:
pi update --self
Only update installed Pi packages:
pi update --extensions