Przejdź do treści
Biblioteka skilli
AI i agenci · Przegląd i audyt

reflect

/reflect

Uruchamia równoległe przeglądy rozmowy, zbiera wnioski i kieruje każdy z nich do konkretnej zmiany w istniejącej umiejętności.

Repozytorium
orca-pstack ↗
Licencja
MIT
Ostatnia zmiana repozytorium
05.11.2026

Ocena i bezpieczeństwo

100/ 100Klasa: dobry
Format25 / 25
Frontmatter z nazwą i opisem, poprawna nazwa, opis do 1024 znaków, nazwa zgodna z folderem.
Odkrywalność25 / 25
Opis mówi, kiedy użyć skilla, ma sensowną długość i jest w 3. osobie.
Treść35 / 35
Co najmniej 150 słów i do 500 linii, nagłówki, listy kroków, przykłady albo kod.
Higiena15 / 15
Bez zaślepek typu TODO, bez ścieżek Windows, z licencją.
Ryzykobrak sygnałów w skanie
Pliki w paczce
5
Skrypty
0
Komentarze HTML
0
Adresy URL w SKILL.md
0
Przestarzałość

Brak sygnałów. Repozytorium zmienione 05.11.2026, bez odwołań do wycofanych modeli i API.

Licencja

MIT, otwarta: można używać i zmieniać, z zachowaniem informacji o autorze.

Co robi i kiedy

Uruchamia równoległe przeglądy rozmowy, zbiera wnioski i kieruje każdy z nich do konkretnej zmiany w istniejącej umiejętności.

Oryginalny opis (EN)

Spawn three parallel review subagents over the active transcript, surface learnings, and route each to a concrete edit on an existing skill. Use when the user says reflect.

Jak działa

  1. 1. Locate the active transcript
  2. 2. Spawn three reviewers in parallel
  3. 3. Synthesize
  4. 4. Structural enforcement check
  5. 5. Apply
  6. 6. Summarize for the user
Sekcje z nagłówków SKILL.md, w oryginale · 77 linii, 1 025 słów
Podgląd SKILL.md
---
name: reflect
description: Spawn three parallel review subagents over the active transcript, surface learnings, and route each to a concrete edit on an existing skill. Use when the user says reflect.
disable-model-invocation: true
---

# Reflect

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

Invoke when the user says "reflect" or "/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.

## 1. Locate the active transcript

Before fanning out, find this conversation's transcript in the active workspace's session directory. Do not glob across every project's directory. That crosses workspace boundaries and reads private chats from unrelated projects.

Where the transcript lives depends on your harness:

- **Cursor:** the `agent-transcripts/` directory the system prompt names.

  ```bash
  ls -t <agent-transcripts>/*.jsonl <agent-transcripts>/*/*.jsonl <agent-transcripts>/*/subagents/*.jsonl 2>/dev/null | head -10
  ```

  That covers the legacy flat (`<id>.jsonl`), current nested (`<id>/<id>.jsonl`), and subagent (`<parent>/subagents/<child>.jsonl`) layouts. For each candidate, read the first JSONL line. Take the file whose `message.content[0].text` contains the conversation's opening user prompt.
- **Claude Code:** `~/.claude/projects/<slug>/<session-id>.jsonl`, where `<slug>` is the workspace path with every character that isn't a letter or digit turned into "-".
- **Pi:** `~/.pi/agent/sessions/--<slug>--/*.jsonl`, where `<slug>` is the workspace path with the leading slash dropped and each "/" turned into "-".
- **Codex:** `~/.codex/sessions/YYYY/MM/DD/rollout-*.jsonl`. Keep only files whose first line has `payload.cwd` equal to the workspace path.
- **Other harnesses:** check the harness's session directory for this workspace.

Outside Cursor, take the newest candidates first and confirm one by finding the conversation's opening user prompt in its first user message. If none matches, pass a tight digest of the session instead.

## 2. Spawn three reviewers in parallel

One message, three `Task` calls, `subagent_type: generalPurpose`, agent mode (`readonly: false`). When **orca-pstack** dispatches a reviewer, start all independent Orca workers and local reviewers, then end the turn, as that skill says. Reviewers need source evidence for the tickets, chat threads, and observability traces the transcript references. Cursor readonly mode may restrict MCP access; other runtimes follow their actual tool schema. Apply **orca-pstack**'s capability manifest and preflight: a worker either retrieves the evidence through its own authorized tools or receives attributable evidence from the coordinator. Mark supplied evidence and inaccessible sources explicitly; do not claim an independent lookup when none ran.

**Other harnesses.** Read **orca-pstack** and its `references/runtime-adapter.md` before spawning. The tool names and fields above describe Cursor; map the required model, capabilities, workspace, and lifecycle against this session's actual tool schema. A missing required capability routes through Orca or is reported unavailable; do not just drop the field. Native results and Orca `worker_done` reports feed the same workflow, but use their own lifecycle operations.

Each reviewer and the synthesizer name a role line in the pstack settings file (`~/.agents/pstack-models.md`). Read that file. No harness loads it for you. Each line names a default. Resolve the line with the **orca-pstack** skill before you set `model`. A value with spaces is `<model> <effort>` or `<agent> <model-id> <effort>`, not one slug. Use the default slug only when the file or the line is missing. Leave `model` unset when the value is `auto` or `inherit-parent`. If this harness cannot run that value, dispatch it with the **orca-pstack** skill. Do not switch family first. If dispatch is impossible, report the seat unavailable under **orca-pstack**; do not substitute another model. Pick each subagent's reasoning level and fast or standard speed for the job, per the Subagents section of the **poteto-mode** skill.

| Lens | Role line | Default `model` | Prompt template |
|---|---|---|---|
| Judgment | `reflect judgment, divergent, synthesizer` | `claude-opus-5-5-xhigh` | `references/judgment-reviewer.md` |
| Tooling | `reflect tooling` | `grok-4.7-xhigh-fast` | `references/tooling-reviewer.md` |
| Divergent | `reflect judgment, divergent, synthesizer` | `claude-opus-5-5-xhigh` | `references/divergent-reviewer.md` |

Pass each template verbatim, substituting the transcript path or digest where marked. Collect findings through the native result mechanism or Orca `worker_done`, according to each reviewer's route. Wait for all required reviewer results before synthesis.

## 3. Synthesize

One `Task` call, `subagent_type: generalPurpose`, `model` from the `reflect judgment, divergent, synthesizer` line (default `claude-opus-5-5-xhigh`), agent mode (`readonly: false`). It spot-verifies citations using verified tool access or explicitly labeled coordinator-supplied evidence, following the same capability checks as reviewers. Pass `references/synthesizer.md` verbatim, with each reviewer's full output inlined where marked. It returns an Accepted / Rejected / Backlog list.

## 4. Structural enforcement check

Move any Accepted item that a lint rule, script, metadata flag, or runtime check would enforce more reliably to Backlog. See the **encode-lessons-in-structure** principle skill.

## 5. Apply

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

File each Backlog item to your team's devex or backlog tracker without waiting. Only the Accepted list waits for approval.

Never edit a skill under `plugins/cache/`. Before approval, show its destination: `.cursor/skills/<skill>/` or `~/.cursor/skills/<skill>/` in Cursor, or your harness's project or user skill folder, such as `.claude/skills/<skill>/` or `~/.agents/skills/<skill>/`. After approval, copy the skill folder there and edit the copy. Never open a PR or push.

Follow each approved row's Routing exactly:

- Trivial existing-skill edit (a one-line bullet, a tightened sentence, a stale fact corrected): the parent does it directly.
- Substantive existing-skill edit (a new section, a new pattern table, more than ~10 lines): hand to your harness's skill-authoring skill and run its draft / test / iterate loop. That is `create-skill` in Cursor (built in) or Anthropic's `skill-creator`. If you have neither, follow the Agent Skills format at agentskills.io. "`create-skill`" below means whichever of these you use.
- `tune description: <skill path>` (the skill exists but didn't trigger when it should have): hand to `create-skill` and run its description-optimization loop.
- `new skill via create-skill: <kebab-name>`: hand creation to `create-skill`. Do not invent the shape ad hoc.

If your environment ships a SKILL.md validator, run it on every touched skill before declaring done.

## 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.
Pliki w paczce (5)
  • SKILL.md 7 KB
  • references/divergent-reviewer.md 4 KB
  • references/judgment-reviewer.md 3 KB
  • references/synthesizer.md 4 KB
  • references/tooling-reviewer.md 5 KB

Wersje

RepozytoriumWynikLicencja
Ta strona · monarchjuno/orca-pstack100MIT
iliaal/ai-skills100MIT
chhoumann/t3-pstack100MIT
chmonitor/chmonitor100GPL-3.0
backnotprop/pstack100MIT
guangshinhaha/shinstack100MIT
unsady/den100Apache-2.0
akdevsaha-dev/factory-skills100MIT
kriegcloud/beep-effect100Apache-2.0

Migawka bazy z 13.11.2026 · wersja paczki c6a7b0768b31 (skrót SHA-256 całej paczki)

Pobrania w ostatnich 30 dniach: 0 · ranking

Zgłoś problem z tym skillem: szkodliwa treść, błędny opis albo prośba autora o usunięcie.

Podobne skille

  • 100meditation
    Prowadzi refleksję po sesji: uruchamia trzech równoległych recenzentów nad transkryptem Claude Code lub Codex, wyłapuje trwałe wnioski i kieruje każdy do konkretnej poprawki istniejącego skilla, czekającej na zatwierdzenie.
  • 90trove-reflect
    Przegląda bieżącą sesję pod kątem trwałych wniosków i kieruje każdy w konkretną zmianę: w skillu albo w notatce, bez automatycznego zastosowania.
  • 90agents-reflect
    Po sesji przegląda, co poszło dobrze i źle, i zamienia każdą lekcję w konkretną zmianę w skillu albo w linii AGENTS.md.
  • 100claude-reflect
    Zbiera poprawki zgłaszane podczas sesji w samouczący się system i przypomina o uruchomieniu /reflect, aby zaktualizować CLAUDE.md o wnioski na przyszłe sesje.
  • 90reflect-yourself
    Prowadzi samouczący się system, który zbiera poprawki, odkrywa wzorce pracy i synchronizuje wnioski ze skillami i regułami, przy kończeniu sesji i po poprawkach.

Uwagi i zgłoszenia

Wiadomość trafi do Bartosza Sokolińskiego. Adres e-mail jest potrzebny tylko do odpowiedzi. Szczegóły w polityce prywatności.