proguha/pstack-pi/reflect

Reflect

On Pi, read the platform mapping, including its per-skill notes, before following this skill.

Mine the current conversation for durable learnings, then route them into skill edits.

When to invoke

Invoke when the user says "reflect" or "/pstack:reflect". Skip when the conversation is trivial, off-topic, or already covered by an existing skill the parent followed correctly. One-offs are not learnings.

Process

1. Locate the active transcript

The parent finds its own transcript before fanning out. It is the current Pi session file, which Pi exposes to every shell command as $PI_SESSION_FILE. Search only this workspace's sessions, never other projects' directories under ~/.pi/agent/sessions/. That reads private chats from unrelated projects.

Render it with the transcript tool at skills/poteto-mode/scripts/pi-transcripts.mjs under the installed plugin:

node <plugin>/skills/poteto-mode/scripts/pi-transcripts.mjs digest "$PI_SESSION_FILE" > <tmp>/transcript.md

A raw Pi session carries large system-prompt records, so reviewers read the digest: user and assistant text, every tool call with its arguments, and truncated tool results, each under its entry number. If $PI_SESSION_FILE is unset, pi-transcripts.mjs find "<opening prompt fragment>" prints this workspace's newest session whose opening user message carries the fragment. If both fail, write a tight digest of the session and pass that instead.

2. Spawn three reviewers in parallel

One message, three Agent calls, subagent_type: "general-purpose", with model set as below. Reviewers need MCP access for context lookups (tickets, chat threads, observability traces referenced in the transcript). Pick a subagent_type that retains MCP access. The prompt forbids file writes. The parent applies edits.

Each reviewer and the synthesizer name a role line in pstack-models.md and a default in Models. Set model to that line's value, or to the default if the sheet or the line is missing. Leave model unset when the value is auto or inherit-parent. If the Agent tool rejects a slug, use the default and say so. If it rejects the default, use the closest valid slug of the same family from its error message.

Pass each template verbatim, substituting the transcript path or digest where marked. Reviewers return findings in the Agent response body.

3. Synthesize

One Agent call, subagent_type: "general-purpose", with model from the reflect judgment, divergent, synthesizer line (default in Models). Pick a subagent_type that retains MCP access. The synthesizer's quality check includes spot-verifying citations, which can require MCP access. Use references/synthesizer.md verbatim, with each reviewer's full output inlined where marked. The synthesizer returns a structured Accepted / Rejected / Backlog list.

4. Structural enforcement check

Sanity-check the synthesizer's Accepted list. For any item that would be enforced more reliably by a lint rule, script, metadata flag, or runtime check, move it from Accepted to Backlog. See the encode-lessons-in-structure principle skill.

5. Apply

Before applying any Accepted edit, present the synthesizer's full Accepted/Rejected/Backlog output to the user and wait for explicit approval. The user picks which subset to apply and may redirect routings. Skill changes affect every future agent in the org. Do not auto-apply.

Backlog items file to whatever devex / backlog tracker your team uses automatically. Only the Accepted list waits for approval.

For each approved Accepted item, follow the Routing field exactly:

• Trivial existing-skill edit (a one-line bullet, a tightened sentence, a stale fact corrected): parent does directly.
• Substantive existing-skill edit (a new section, a new pattern table, more than ~10 lines): hand to the skill-development skill and run its draft / test / iterate loop.
• tune description: <skill path> (the skill exists but didn't trigger when it should have): hand to skill-development and run its description-optimization loop.
• new skill via skill-development: <kebab-name>: hand creation to skill-development. Do not invent the shape ad hoc.

If your environment ships a SKILL.md validator, run it on every touched skill before declaring done. Skip this step if it doesn't.

6. Summarize for the user

Short list, no preamble:

• Edits applied: <skill path>. What changed, one line each.
• New skills created: <skill path>. One line each (rare).
• Backlog filed to the devex tracker: <issue title> (<tags>). One line each.
• Dropped: one line per rejected finding + reason from the synthesizer.

Models

Role defaults, stamped by pstack-pi's scripts/sync-pstack.mjs from upstream models.json and the Pi model map in port.json. A matching role line in ~/.pi/agent/pstack-models.md overrides each at runtime; /pstack:setup-pstack writes it.

• reflect tooling: openai-codex/gpt-6-sol
• reflect judgment, divergent, synthesizer: openai-codex/gpt-6-sol

Reasoning effort

A role value in the override sheet may name a reasoning effort after its model, as in openai-codex/gpt-6-sol @xhigh. Levels on Pi: low, medium, high, xhigh, max. Which ones apply depends on the model. A value without @ takes the sheet's default effort line, a level or session, and session when the sheet has no such line. session sets no effort, so the dispatch is the usual one. Strip the suffix before reading the model: inherit-parent or auto still omits model at every level, and a model name is passed as model. On Pi, pass the level as the Agent tool's thinking field and keep the usual subagent_type and model. Pi clamps a level the model does not support.