tobihagemann/turbo / claude/skills/self-improve
self-improve
VerifiedExtract lessons from this session or past sessions and route them to instructions, memory, or skills. Use when the user asks to "self-improve", "what did we learn", or "capture lessons".
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 4.8 KB |
| references/past-session-sweep.md | markdown | 951 B |
| references/transcript-miner.md | markdown | 13.0 KB |
SKILL.md
View on GitHub---
name: self-improve
description: "Extract lessons from this session or past sessions and route them to instructions, memory, or skills. Use when the user asks to \"self-improve\", \"what did we learn\", or \"capture lessons\"."
---
# Self-Improve
Review the current conversation, or when asked the project's past sessions, to extract durable lessons, then route each one to the right knowledge layer.
## Step 1: Gather Session Evidence
### Recover Pre-Compaction Evidence
**Skip** when the conversation is visible in full from the user's own first message, or when the summary shows that the last work before the compaction was a finished `/self-improve` run.
Otherwise, when it starts from a summary of earlier work, recover the compacted turns from the on-disk transcript. Spawn a single subagent (`model: "opus"`, no `name`). Wait for it to report before continuing; do not relaunch it if it has not yet reported. The subagent's prompt must include:
1. The absolute path of the project root
2. A distinctive phrase from the visible conversation, for confirming which transcript belongs to this session
3. An instruction to read `${CLAUDE_SKILL_DIR}/references/transcript-miner.md` for transcript location, extraction, and output format
Treat the returned items as raw evidence for the scan below.
### Sweep Past Sessions
**Run** when asked to distill sessions beyond the current one: read [references/past-session-sweep.md](references/past-session-sweep.md) and follow it. **Skip** otherwise.
## Step 2: Scan for Lessons
Scan the full conversation with this priority:
1. **Corrections** — Where the user interrupted, said "no", "actually", "stop", "not like that", redirected, reverted a change, picked a variant other than the one Claude proposed, or manually fixed something Claude did wrong, including a value the user changed in a file Claude wrote. After a run of nudges toward one target, the lesson is where the user settled, not each step. Highest-value lessons.
2. **Repeated guidance** — Instructions the user gave more than once. Across separate sessions this counts even where each instance reads as ordinary steering on its own.
3. **Skill-shaped knowledge** — Domain expertise that was needed repeatedly, tool/API integration details that had to be looked up, decision frameworks that emerged for evaluating options, content templates or writing conventions that were refined, and multi-step workflows where ordering mattered (as reusable domain knowledge, not the workflow itself — see #4).
4. **New workflows** — Did this session establish a novel multi-step procedure, coordination pattern, or automation that worked? A successful workflow that would need to be repeated is a prime skill candidate — even if it ran fine this time. Distinct from #3: this captures the procedure itself as a repeatable artifact, not knowledge about how to do it. Flag it.
5. **Preferences** — Formatting, naming, style, or tool choices the user expressed.
6. **Failure modes** — Approaches that failed, with what worked instead. For tool or script call failures, trace back to the information source that led to the error and make fixing that source the lesson (e.g., clarify a reference file, update skill instructions, add missing documentation). When the failure is a mistake in code that an automated check could have caught, read the project's check commands and CI configuration, once per run, before framing the lesson: a check that exists but that no git hook or CI job runs is the lesson.
7. **Domain knowledge** — Facts or conventions Claude needed but did not have.
8. **Improvement opportunities** — Out-of-scope improvements noticed during work: code that could be refactored, missing tests, performance issues, readability concerns, or feature ideas that were intentionally skipped to stay focused. **Skipped findings count here**: when code simplification or code review identified a genuine improvement or issue but it was skipped for this session, record it as a project improvement so it isn't lost.
9. **Trusted reviewer feedback** — Human PR review comments that reveal project conventions, patterns, or corrections. Trusted reviewers are repo collaborators with `admin` or `maintain` roles (determine via `gh api repos/{owner}/{repo}/collaborators --jq '.[] | select(.role_name == "admin" or .role_name == "maintain") | .login'`). Their feedback takes precedence over other reviewers and AI bots when there are contradictions.
Record each lesson with the evidence behind it, its origin (this session, or the past session it came from), the skill whose workflow it clearly arose from when there is one, and any reason the user did not state.
## Step 3: Run `/route-lessons` Skill
Run the `/route-lessons` skill with the lessons from Step 2. When the scan found none, say so instead.
Then use the TaskList tool and proceed to any remaining task.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |