pi-result-handling — Claude Code skill: what it does, how to install
Skill for presenting Pi helper output correctly in Claude Code, Cursor and other IDEs. Preserves review structure, file paths and line numbers as reported.
npx skills add Agents365-ai/365-skills --skill pi-result-handling -a claude-codeWhat pi-result-handling does
The skill tells the agent how to present Pi helper results. Review findings are ordered by severity, file paths and line numbers are kept verbatim. When Pi marks something as an inference or open question, that distinction is preserved. If there are no findings, the agent states that explicitly and keeps the residual-risk note short.
After a review, the agent will stop and show the list of findings without making changes. To apply fixes, you must specify which issues to address. Auto-fixing is forbidden, even when the bug is obvious. If Pi returns an error or fails to start (missing binary, no provider configured), the agent will show the most actionable stderr lines and suggest /pi:setup.
The skill is not user-invocable — it activates in the background when other Pi commands return data.
What it helps with
- preserve Pi helper output structure and finding order by severity
- keep file paths and line numbers exactly as Pi reports them
- distinguish facts, inferences and open questions in the report
- stop the agent after review, preventing code edits without permission
- surface diagnostics when Pi fails to start or returns an error
Example prompts for the agent
- "Run a Pi review on this file" — the skill takes Pi output and shows findings ordered by severity
- "Check the project with Pi helper" — after the review completes, the agent stops and asks what to fix
- "Fix the second bug from the Pi list" — the skill passes the command, and only the named bug will be touched
How to install pi-result-handling
1Skills CLI (Vercel)
npx skills add Agents365-ai/365-skills --skill pi-result-handling -a claude-code2Manually: put the files in place
~/.claude/skills/pi-result-handling/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/pi-result-handling/
1Skills CLI (Vercel)
npx skills add Agents365-ai/365-skills --skill pi-result-handling -a codex2Manually: put the files in place
.agents/skills/pi-result-handling/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add Agents365-ai/365-skills --skill pi-result-handling -a cursor2Manually: put the files in place
.cursor/skills/pi-result-handling/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install Agents365-ai/365-skills pi-result-handling2Manually: put the files in place
.github/skills/pi-result-handling/SKILL.md1Skills CLI (Vercel)
npx skills add Agents365-ai/365-skills --skill pi-result-handling -a gemini-cli2Platform CLI
gemini skills install https://github.com/Agents365-ai/365-skills --path plugins/pi-plugin-cc/skills/pi-result-handlingManually: put the files in place
.devin/skills/pi-result-handling/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add Agents365-ai/365-skills --skill pi-result-handling -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about pi-result-handling
How to install pi-result-handling in Claude Code?
Run npx skills add Agents365-ai/365-skills --skill pi-result-handling -a claude-code, then the path ~/.claude/skills/pi-result-handling/SKILL.md will be ready.
What does pi-result-handling do?
The skill shows the agent how to format Pi helper output correctly: preserve review structure, file paths and line numbers, distinguish facts from inferences. After a review, the agent stops and does not edit code without permission.
Can I invoke pi-result-handling directly?
No. The skill runs in the background and activates automatically when other Pi commands return data to present to the user.
