Skip to content

mikeparcewski/wicked-garden / skills/learn

wicked-garden-learn

Verified

Learn from outside sources, every claim cited. Learn mode: research a topic or event, synthesize a claims ledger (each claim with its source URL and observed_at), walk the operator through it, then author .wicked/knowledge/<slug>/ plus text-only skills that cite the claims; a refresh proposes only changes (confirm, supersede, contradict, retire), never a silent delete. Report mode (the `sourced-report` workflow): the same research and synthesis, without the write-up, from staged documents, the job's picked sources and the web, then a fidelity check by a seat that did not gather. It changes nothing. Use when: "learn about X", "research X and save what we learn", "what happened with <event>", "refresh the knowledge on X", "check the learned knowledge", a brief with a `Lens:` line, or a phase of the `learn` or `sourced-report` workflow. NOT for this repository's own code (wicked-garden-repo-learn) or for what the operator told you (a memory: wicked-garden-mem).

FileKindSize
SKILL.mdskill md10.7 KB
refs/claims.mdmarkdown2.7 KB
refs/refresh.mdmarkdown2.8 KB
refs/report-analyze.mdmarkdown1.2 KB
refs/report-check.mdmarkdown4.4 KB
refs/report-claims.mdmarkdown7.1 KB
refs/report-competitive.mdmarkdown1.2 KB
refs/report-critique.mdmarkdown1.3 KB
refs/report-due-diligence.mdmarkdown1.3 KB
refs/report-estimate.mdmarkdown1.4 KB
refs/report-research.mdmarkdown1.1 KB

refs/claims.md

View on GitHub
# The claims ledger

`claims.jsonl` holds one JSON object per line. The ledger is the record. README.md and the skills
are views over it, and both cite claims by id.

## The record

```json
{"id":"c-1f3a9b2e",
 "text":"In WAL mode, readers do not block writers and a writer does not block readers.",
 "sources":[{"url":"https://sqlite.org/wal.html","title":"Write-Ahead Logging","published_at":null,"observed_at":"2026-10-11T04:10:00Z"}],
 "status":"active",
 "first_seen":"2026-10-11T04:10:00Z",
 "last_verified":"2026-10-11T04:10:00Z"}
```

| Field | Rule |
|---|---|
| `id` | `c-` followed by the first 8 hex characters of the sha256 of the claim text, lowercased with whitespace collapsed, taken at first sight. The id never changes. A reworded claim is a new claim, and the old one becomes `superseded`. Compute it with `wicked-garden run scripts/learn/check.py --claim-id "<text>"`. |
| `text` | One atomic, checkable statement, at least 40 characters. An `event` topic's claim states its date. |
| `sources` | At least one `{url, title, published_at, observed_at}`. `url` is `http(s)`. `observed_at` is the UTC time you fetched it, in ISO-8601. |
| `status` | `active`, `superseded`, `contradicted` or `retired`. |
| `reason`, `evidence` | Required when the status is not `active`. `evidence` is `[{url, observed_at}]`, the source that shows the change. |
| `superseded_by` | Required for `superseded`: the id of the replacing claim. |
| `first_seen`, `last_verified` | `last_verified` moves forward each time a refresh re-sees a source that still supports the claim. |

## README.md

- One short section per question from scope.
- Every factual sentence ends with its claim ids: `… readers never block writers [c-1f3a9b2e].`
- Every active claim is cited at least once.
- The README cites no claim that is not active.
- `## Sources` lists each source once: URL, title and observed date.
- `## Operator notes` holds what the operator told you in the walkthrough. Each note is labelled
  "operator, <date>". These notes are memories, not claims, so they carry no claim id.

## A skill from the ledger

```markdown
---
name: sqlite-wal-setup
description: |
  Configure SQLite write-ahead logging for a service with many readers and one writer:
  enable WAL, set the checkpoint policy, and know what WAL cannot do on a network filesystem.
metadata:
  learn-topic: sqlite-wal
---

# SQLite WAL setup
1. Run `PRAGMA journal_mode=WAL;` once per database file. The mode persists [c-5b0e77a1].
…

## Sources
c-5b0e77a1, c-1f3a9b2e (see .wicked/knowledge/sqlite-wal/claims.jsonl)
```

Then run `wicked-garden run scripts/learn/check.py --repo .`. If it exits 0, there are no findings.
If it exits 1, each finding names its file, its rule and the fix. Exit 2 means there is no topic.
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: 472kUpdated Claude CodeCodexVerifiedInstall
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: 322kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 322kUpdated Claude CodeCodexVerifiedInstall
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 availableStars: 322kUpdated Claude CodeCodexVerifiedInstall