compile-knowledge · git:20261023.a3e54d5 · 2026-10-23 · sha256 557c61dafd973d3c
compile-knowledge git:20261023.a3e54d5A
Immutable. This exact content never changes and is served at /api/v1/blob/557c61dafd973d3c.
No tool to install first
claude plugin marketplace add https://markdownregistry.com/mp/a/art_dwwpde6rrwahckhw/557c61dafd973d3c8a910e2336faa4690425d976124e01ad58152d31b6d974a2.json
Then /plugin install compile-knowledge@markdownregistry-dwwpde6rrwahckhw in Claude Code. It installs this exact version, the one graded A here, pinned to the commit it was graded at, and leaves it there.
---
name: compile-knowledge
description: >-
Compile durable, judgement-shaped knowledge into interlinked markdown —
atomic files, [[wiki-links]], a maintained index. Use when persistence is
asked for explicitly ("save this", "write it to the wiki") or when the work
produced a decision record, a cause, or a gap analysis; ordinary factual
answers and routine task results are not a trigger.
---
# compile-knowledge
## When this fires
Two triggers, and only these:
1. **An explicit persistence ask** — "save this", "write this to the wiki",
"update the wiki/memory", "log this finding", "structure this knowledge",
"follow the karpathy method".
2. **Durable judgement-shaped output** — a decision record, the CAUSE behind a
finding, a gap analysis, a non-obvious fact that cost real work to learn and
would cost it again.
It does **not** fire on ordinary analytical work. Answering a factual question,
summarising a file, explaining code, or reporting a routine task result
produces nothing durable to compile. If the only thing you could write down is
a restatement of what the repository or the transcript already says, skip it.
Durable knowledge is worth keeping as many small, interlinked markdown files,
compiled over time and surfaced through an index — not as one giant doc, a chat
log, or a one-off file that rots. This skill makes compiling consistent so your
agent gets smarter over time instead of relearning the same things.
## Where it goes — pick the right store
- **Agent memory** (default, always available): your `.claude/.../memory/`
folder with `MEMORY.md` as the index. This is the per-agent store and it
survives restarts — it's the karpathy "external memory" that keeps you sharp
across sessions. Governed by the memory rules already in your system prompt —
follow them. For most agents this is the only store you need.
- **Team wiki** (when other agents share your box): what the whole team should
know about the owner's business — customers, prices, suppliers, decisions, how
the owner likes things done. Publish it with
`5dive memory add --store=wiki --name=<slug> --description="<one line>"`
(the body on stdin); that writes the file and its index line, and every
colleague's `5dive memory search` finds it. Don't keep a `wiki/` folder in your
project instead: no other agent's search reads it. Skip this if you are the
only agent; don't manufacture team ceremony.
Rule of thumb: "only I act on this" → memory. "Anyone on my team might need
this" → team wiki. Cross-link between them with `[[slug]]` when they relate.
## The async pipeline vs. you — division of labor
Some platforms run an automatic consolidation pass over finished sessions (on
5dive: `5dive memory consolidate`, scheduled for you by the heartbeat). If your
platform has one, know what it covers and what only you can do:
- **AUTOMATIC — plain facts.** The pass distils your FINISHED session
transcripts into memory atoms in your own store. It never reads the live
session, and nothing it writes leaves your box. You do NOT need to hand-copy
plain facts out of a session to keep them — that is what stops knowledge
dying with the context window.
- **STILL YOURS — judgement.** A wiki page, a decision and its reason, a gap
analysis, the CAUSE behind a finding: the pipeline can only lift what is
stated in the transcript, and a conclusion you drew is not lying there. It
also cannot publish to a shared wiki — it has no store selector, and shared
stores are deny-default by design — so publishing to the team store stays a
curated act, and that act is yours.
This narrows what you compile; it does not replace compiling. Judgement-shaped
knowledge still goes through the procedure below, before you close the task.
## Before you write — the hygiene gate
Compile ONLY a durable, non-obvious fact. Skip and move on if it is:
- routine / derivable from the repo, git history, or existing docs,
- true only for this one conversation,
- already covered by an existing file (→ UPDATE that file instead, don't duplicate).
Most tasks (a deploy, a restart, a one-line fix) produce nothing durable. That is
fine — do not manufacture a memory to "have written something." Filler is worse
than nothing; it pollutes recall.
## The procedure
1. **Search first.** Look for an existing file on this topic (on 5dive,
`5dive memory search "<topic>"` reads your store and the team wiki). If one
exists, edit it — never create a near-duplicate.
2. **Atomic.** One fact / one topic per file. If you're tempted to add a second
unrelated fact, that's a second file.
3. **Name it.** kebab-case slug, with a type prefix for memory
(`feedback_…`, `project_…`, `reference_…`, `user_…`) or a clear topic slug for
the wiki. The slug is the link target.
4. **Frontmatter.** `name` (= the slug), `description` (ONE line — this is what
gets matched during recall, make it specific), and a `type`/category.
5. **Body.** State the fact plainly. Link related entries with `[[slug]]` —
liberally; a link to a file that doesn't exist yet is a fine TODO marker. For
`feedback`/`project`, follow with **Why:** and **How to apply:** lines.
6. **Index.** Add or update a ONE-LINE pointer in the index (`MEMORY.md` for
memory; `5dive memory add` writes the team wiki's line for you):
`- [Title](slug.md) — hook`. Keep it under ~200 chars; detail lives in the
file, never the index.
7. **Hygiene.** Delete files that turned out wrong. Convert relative dates to
absolute. If the index is getting long, tighten lines — don't let it bloat.
## Lifecycle — supersede, don't silently overwrite
Facts rot. When one is time-sensitive, uncertain, or replaces an older one,
stamp the envelope so recall can age it out instead of surfacing stale truth:
- **`valid_to: YYYY-MM-DD`** — date the fact expires / needs a recheck. Past it,
recall keeps the fact but demotes + flags it `⚠ expired`.
- **`supersedes: <slug>`** — when a new fact replaces an old one, point at the
old slug. The old fact is then demoted (`⤴ superseded`) in recall instead of
lingering as a second, contradictory answer. Prefer this over edit-in-place
when the OLD value is still worth seeing (audit trail); edit in place when it
isn't.
- **`confidence: high|medium|low`** — low/medium facts are demoted so a hunch
never outranks a verified fact.
- **`provenance: "<source>"`** — where the fact came from, distinct from
`compiled_by` (who wrote the note).
If your store has a CLI, it likely exposes these as flags — on 5dive, `5dive memory add` takes
`--valid-to=`, `--supersedes=`, `--confidence=`, `--provenance=`. All optional — omit them and
behaviour is unchanged, and the frontmatter fields above are the portable part.
## Anti-patterns
- A wall-of-text doc instead of atomic files.
- Research left as a standalone `notes.md` that never gets folded in — that's
working notes, not knowledge. Compile the durable parts into the store.
- Duplicating a fact across memory AND wiki — pick one home, cross-link.
- Index entries that restate the whole file.
- Writing filler to satisfy a habit/checklist.
## Quick checklist
`[ ] durable & non-obvious? [ ] right store? [ ] updated existing vs new?`
`[ ] atomic + named + frontmatter? [ ] [[links]]? [ ] index line added?`