Schema 1.8, after Google's Self-Improvement Memory paper (arXiv 2611.18856).
- Error book: one entry per recurring problem with every attempt, its
expected and observed result and context; aliases on merge (ckb-lint §9).
- Decisions: expected_outcome, outcome, outcome_checked, ## Tried,
append-only ## Outcome, an outcome-review procedure (ckb-decide); lint
validates outcomes and closes a due review once outcome_checked >= review_on.
- Lessons: Context line, upstream lessons from linked/ in the session
brief, adopt and merge rules (ckb-profile, session_start.py); ckb-init
Step 8b carries lessons and problems into a new KB; ckb-upgrade b5.
- Deferred: a self-improvement loop over the template. Rejected: loading
the whole record into context.
No VERSION change or changelog row: made on the merge from dev (D-0022).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
All work lands on dev; VERSION and the changelog change only when dev is
merged into experimental, test or main. Paths: dev->experimental->test->main,
dev->test->main, dev->main. A version keeps its number when promoted; main
is merged back into dev, test and experimental after a release. 1.x keeps
its own fixes. experimental becomes the preview channel; dev is not a
channel, and ckb-init/ckb-upgrade offer experimental when asked for it.
README, manuals, changelog channel sections and D-0022 updated; Polish
manual gains the profile-skills section.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- 2.1.0 new opt-in `voice` module (D-0020): skill ckb-voice rewrites every
prose file written under apply_to (outputs/answers, documents, reports,
PROJECT-OVERVIEW) in a voice whose profile is a wiki page,
wiki/voice/<name>.md (type voice-profile); settings in ckb.yaml (mode,
profile, writer, apply_to, skip, chat); frozen spans never change, vague
claims flagged not cut; `writer:` delegates to an external rewriting
skill such as a personal ghost-writer. ckb-module writes a manifest's
default_config into the entry's config:. Eval ckb-module/install-voice.
- 2.2.0 new opt-in `convert` module (D-0021): skill ckb-convert and
convert.py turn Markdown into Word/ODT/PDF/HTML through pandoc with a
house style in templates/styles/; unit tests and eval
ckb-module/install-convert.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Fixes F1-F5 from the 2.0.0 script review:
- removeprefix/removesuffix (3.9+) replaced; lint, retrieve and
consolidate crashed on Python 3.8 with any graph edges
- routines.py --mark wrote `last_run:DATE` (invalid YAML) and swallowed
trailing comments into the value
- export_okf.py --out must be inside outputs/ (it is rmtree'd);
export_starlight.py image copies confined to wiki/assets/
- script counts in AGENTS.md, READMEs and manuals
New tests/unit/ (stdlib unittest, run from smoke.sh): 3.8 API guard,
routines round-trips, exporter containment, AGENTS.md consistency.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Breaking: lint_quick.ps1 and every PowerShell path are removed. Without
Python the session-start fallback is lint_quick.sh (Git Bash on Windows).
Every script the template and its modules ship is Python. KBs that need
PowerShell stay on the 1.x maintenance branch. ckb-upgrade now deletes
files upstream removed from a skill or module folder it still ships.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Adds the `1.x` release channel (fixes only, last line with PowerShell
support) to every channel table, and makes ckb-upgrade stop and ask before
crossing a major version, offering the maintenance branch instead.
D-0019 records the 2.0 plan: drop PowerShell, scripts stay in Python.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
New opt-in `share` module with skill `ckb-share` and stdlib `share.py`
(join/send/unshare/list/receive/status). KBs exchange documents through a
git channel with one append-only mailbox per KB; sends are packed
self-contained, secret-scanned and confirmed; received documents land in
raw/inbox/ (untrusted), retractions in raw/retractions/; the git author
must own the mailbox. dedup_inbox.py reports SAME SHARE; ckb-ingest and
ckb-retract know share files. Smoke test covers two KBs end to end.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ABOUT.pl.md ships as the Polish scaffold next to ABOUT.md, like the other
.pl.md docs. Agents read both when both are filled in; ABOUT.md wins on a
conflict (AGENTS.md, ckb-ingest). ckb-init copies both and puts a Polish
answer in the Polish file; ckb-upgrade creates it once for KBs below
1.26.2; ckb-reset leaves it alone; the mcp server's kb_overview includes
it and its description ignores either placeholder. READMEs, QUICKSTART.pl,
both manuals and both changelogs updated.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
New core skill ckb-feynman (plain-words explanation of one page, Wiki /
Sources / Mine gaps, explain-back, numbered next moves, sessions in
outputs/teaching/feynman/). Registered in AGENTS.md routing and
disambiguation, ckb-init, ckb-teach-me (skips the folder, reserves the
slug), both READMEs, manuals and changelogs. outputs/teaching/ added to
the directory contract. tests/run_evals.py matches deleted files with
glob semantics. Eval case tests/evals/ckb-feynman/session.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
docs/QUICKSTART.md and docs/QUICKSTART.pl.md take a new user from an
empty folder to the first ingest, answer, lint and sync. Linked from
both READMEs and both manuals; ckb-init copies them.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- F11 new opt-in module `capture` (skill ckb-capture, capture_url.py):
stdlib fetch with readable-text extraction, Wayback only over a failed
or thinner live page, --from-file for the agent's own fetch, PDFs saved
as files, --media via yt-dlp with consent, untrusted: true marking,
dedup at capture time; dedup_inbox.py now compares source_url (D-0017)
- F22 tests/run_evals.py (list/prepare/check/validate) and five
structural eval cases; reference run for ckb-decide/record
- F18 manual section with untested setups for Codex CLI, Cursor,
OpenCode and Gemini CLI, plus the Claude Code-only feature table;
AGENTS.md tells agents without skill loading to read SKILL.md
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- F13 new skill ckb-consolidate with consolidate.py: a proposals-only
report in outputs/reports/ (topic clusters and their state, merge
candidates, missing links, stale pages, old log entries by month);
never edits a page, never rewrites the log
- F15 topic pages (type: topic) in the new wiki/topics/, built only on
approval, with member-page digests as ## Sources so lint flags drift;
ckb-retrieve routes "overall" questions to wiki/topics/index.md first
- Schema 1.7 (D-0016): ckb-upgrade b4, ckb-init and ckb-reset know
wiki/topics/; ckb-routine suggests a fortnightly Consolidate --confirm
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- F10 bm25 section search: stdlib SQLite FTS5 over page sections, with
title/heading path/tldr in a separate context column; cached in
tmp/ckb-fts.sqlite, refreshed by mtime then sha256; section anchors
per candidate. New benchmark tests/bench_retrieval.py over
tests/fixtures/bench-kb (20 questions): hit@1 14/20, hit@3 20/20 with
and without bm25; weight kept at 1.0 (D-0015)
- F12 token estimate per candidate; layered reading in ckb-retrieve
- F16 --opened access log in tmp/access.log; lint_report.py --cold
proposes demotion to an index `## Archive` section, never moves
- F17 ckb-retrieve offers a table/chart/deck/HTML once when it fits
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Schema 1.6, all fields optional (D-0014):
- F7 review: `review: draft|reviewed|needs_update`, `reviewed_by`,
`reviewed_on`; lint check 17 and a review queue; ckb-lint marks reviews
- F4 `> CONTRADICTION:` marker, lint check 18; validity dates deferred
- F21 `tags` against the new `wiki/tags.md`, lint check 19,
retrieve_candidates.py --tag, DRAFT/NEEDS_UPDATE flags
- F9 profiles in workload/profiles/, protected `## Rules` on project
scope pages, precedence AGENTS.md > Rules > profile
- F8 explicit lessons (profile or wiki/lessons.md); new ckb-profile skill
Carry-over: `.secrets-allow` `<glob> match:<hash>` entries (also in the
bash/PowerShell fallbacks); open threads no longer print twice.
ckb-upgrade migrates 1.5 -> 1.6; init/reset know tags.md; Starlight
shows review and tags. Smoke test covers a 1.6 KB. Docs in EN and PL.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Seven features from the 2026-10-31 agent-knowledge research, no schema
change (kb_schema_version stays 1.5):
- F1 session-start hook: session_start.py --hook, start_hook.py, opt-in
via session.start_hook in ckb.yaml (D-0011)
- F2 kept answers in outputs/answers/, lint check 16 (D-0013)
- F3 ingest check-in before writing; batch mode skips it
- F5 open threads in workload notes, collected at session start (D-0012)
- F6 dedup_inbox.py: duplicates reported before ingest
- F19 lint check 15 for secrets and <private> spans, --secrets before
every sync commit, redaction in saved conversations (D-0013)
- F20 context budget: lint_report.py --budget, budget in ckb.yaml
Smoke test covers every new script. Docs updated in EN and PL.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
announce_title replaces ask_title; no question, no waiting. D-0006 reverses
point 1 of D-0005.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- Rule E opens a new session with the title question (conversations.ask_title);
save_conversation.py --set-title stores it, saved transcripts and workload use it
- New ckb-routine skill + routines.py: time-since-last actions in ckb.yaml,
checked at session start, run after the user's request, last_run committed
- Docs EN/PL, D-0005, smoke test coverage for both scripts
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- ckb-init no longer copies tests/.
- ckb-upgrade drops tests/ from the template layer and offers once to
delete a leftover copy from an older template.
- Directory contract, READMEs, manuals and changelogs (en/pl) say so.
- Smoke test itself unchanged.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- Rule E resolves git user.name/user.email once per session; OS user as
stated fallback.
- Rule D: one workload file per person per day,
YYYY-MM-DD_<author>_summary.md with type: session frontmatter; older
files read as before.
- Rule B log format gains an optional, lint-neutral **Author:** line;
wiki/log.md header and ckb-upgrade example updated.
- save_conversation.py records an author row (--author override); smoke
fixture passes a fixed author.
- ckb-project-summary and ckb-module cite the new workload name.
- D-0004 records the decision; manual, READMEs and changelogs in English
and Polish.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- ckb.yaml: conversations.autosave (default false). On, Rule E saves every
session at its end without asking; ckb-add-conversation toggles it on
request.
- Per-machine half: autosave_hook.py --check/--install/--remove manages a
Claude Code Stop hook in gitignored .claude/settings.local.json. Rule E
runs --check at session start when autosave is on and asks before
changing machine configuration; toggling the switch asks the same.
- save_conversation.py --from-hook reads transcript_path from hook stdin.
- ckb-init writes the block, ckb-upgrade leaves it alone.
- Manual, READMEs and changelog in English and Polish; Polish changelog
also gains the missing 1.11.1 and 1.12.0 rows.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- New skill ckb-add-conversation: on request, copies the session's Claude
Code transcript (JSONL) into raw/saved-conversations/<date>/ with a
readable Markdown rendering and an index.md; answers "did we discuss X"
from those files. Saved conversations stay outside retrieval and ingest.
- AGENTS.md: directory contract, cascade note, routing row, disambiguation,
Rule E session-end ask (save only on a yes).
- ckb-retrieve, ckb-ingest, ckb-reset (tier 3), ckb-init know the folder.
- Docs, both READMEs, CHANGELOG, VERSION; smoke test renders a fixture
transcript against a golden file.
- .gitignore: ideas/ local scratch folder for template ideas.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- D-0003: <python> interpreter resolution, per-OS install guide, lint_quick.sh/.ps1
session-start fallback, manual lint checklist, no-script retrieval path,
exporters declare Python a hard requirement
- Lint check 13 pairs each Crux quote block with the attribution under it, so a
multi-source Crux no longer cross-fails; fixture KB gains a second source
- Session summaries for 2026-09-29 and 2026-10-01
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- Skills: descriptions cut to routing-hint length (1925w -> 840w), Trigger
phrases removed, Purpose collapsed, Edge cases moved to references/.
- Rule B names the decisions-log exception; "Sync the wiki" trigger dropped;
reserved sections "Four" -> "Three"; directory contract adds ckb.yaml,
docs/, tests/.
- MANUAL/CHANGELOG/OPENSPEC (EN+PL) moved to docs/; template ADRs D-0001 and
D-0002 restored from 0c06cb6 into docs/decisions/; referrers repointed.
- ckb-retrieve: rank fusion moved into scripts/retrieve_candidates.py;
Purpose states the one write it owns (query-gaps).
- tests/: fixture KB, golden JSON, smoke.sh.
- Scaffold: overview.md describes content not schema; graph/index.md names
base skills only; query-gaps.md loses its self-rotting freshness window.
- README/MANUAL/OPENSPEC/CHANGELOG ghost-writer pass in both languages.
- VERSION 1.9.0 -> 1.10.0.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The template repo now keeps three branches with fixed meanings — main is
stable, test is the release candidate, experimental is development — and
ckb-init/ckb-upgrade can source from any of them instead of only main.
Selection is per-invocation, in words the user already uses ("initialize
from the test branch", "check experimental for updates", "switch back to
stable"), and sticky: the resolved repo and branch are written to a
template: block in ckb.yaml. Without persistence, a KB bootstrapped from
experimental would be silently pulled back to main by its next upgrade.
A missing file or missing block both mean main, so every KB predating
this convention behaves exactly as before.
One consequence needed explicit handling. A KB tracking test or
experimental can sit on a VERSION main has not released yet, so comparing
it against main finds nothing newer — which the version check would have
reported as "up to date". That is true and misleading. ckb-upgrade now
reports it as "ahead", and treats a move back to main as a downgrade:
explicitly confirmed, with the specific losses named, and blocked
outright where kb_schema_version would drop below what local pages are
already written against.
ckb-module is told not to clobber the template: block — a module install
that silently reset a KB's channel would change what its next upgrade
pulls, which is not a module's business.
Documented in both READMEs, both MANUALs and both CHANGELOGs. VERSION
1.8.0 -> 1.9.0; kb_schema_version stays 1.5, since this is tooling rather
than a content contract.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Two version numbers exist in this project and are easy to confuse:
kb_schema_version (the content contract, in wiki/index.md) and VERSION
(the tooling layer). Until now neither had a written history, and the
page schema was documented in four places at three levels of detail.
CHANGELOG.md / CHANGELOG.pl.md consolidate both: the current schema in
full (frontmatter for all pages and for decisions, the three reserved
body sections, the closed edge vocabulary with a "since" column, the
reserved scaffold), then the schema history 1.1-1.5 and the template
history 1.0.0-1.8.0.
The history is reconstructed from git rather than from memory, so it
records what actually happened rather than a tidied version of it:
- There was never a schema 1.0; versioning began at 1.1 on 2026-07-13.
- Template 1.4.0 and 1.5.0 were never published — VERSION jumped from
1.3.0 to 1.6.0 on 2026-09-01.
- Connector-backed libs shipped as tooling in 1.1.0, but the schema only
recorded them at 1.2 nine days later. The schema column shows what was
in effect after each release, with a footnote on the lag.
Registered with ckb-init (copied verbatim into a new KB) and ckb-upgrade
(taken wholesale from upstream rather than merged, since upstream is
authoritative about its own history). Cross-linked from both READMEs and
both MANUALs.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Graft keeps a derived, disposable code graph in sync with a content hash
rather than a calendar, and keeps a protected block on every regenerated
node. This KB is the opposite kind of store — durable, curated, built from
material that cannot be regenerated — but several of Graft's mechanisms
port cleanly, and two of them close real gaps here.
Schema 1.5 is additive: every 1.4 page remains valid.
1. `## Crux` — verbatim source excerpts alongside the synthesis. A summary
can drift silently; a quote either still matches its source or it does
not. Lets `ckb-retrieve` ground an answer without a round-trip to the
archive, and makes drift mechanically detectable.
2. `## Notes` — human-authored and protected everywhere. Closes a real
gap: `ckb-index-external` regenerates connector pages wholesale, so an
annotation written there was previously destroyed on the next refresh.
3. `source_fingerprint`/`source_checked` — a digest of the material a page
was built from. Freshness by date says a page has aged; a fingerprint
says whether its evidence moved. Most valuable for connector-backed
libs, where documents change with no notice.
4. `lint_report.py --quick` — a deterministic one-line session-start
signal, wired into Rule E next to the existing `git status` check.
5. In-degree as a rank-fusion signal in `ckb-retrieve`, weighted below 1.0:
centrality is a prior, not evidence.
6. Blast radius — a new `ckb-ingest` step walking the graph backwards from
touched entities to find what the incoming material contradicts, before
writing anything. Ingest was additive-first, which is how a wiki
accumulates two pages that quietly disagree.
7. Edge vocabulary in `wiki/graph/index.md` rewritten as a question per
verb, and completed: `part_of` was written by `ckb-code-map` but never
declared. Added `produces`, `configures`, `validates`, `implements`.
Lint gains checks 12 (fingerprint drift), 13 (crux verbatimness) and 14
(the protected-Notes rule), verified against a synthetic fixture covering
stale digests, missing sources, fabricated quotes and paraphrased evidence.
Not adopted: the gitignored regenerable store, the MCP server and CLI
daemon, tree-sitter parsing, statusline hooks, telemetry.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Commit all in-flight work — ckb-module and ckb-reset skills, the
.agents/modules/ scaffold, OPENSPEC docs, decision records D-0001 and
D-0002, graph edges and workload summaries — so the reset that follows
is fully recoverable.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Decision log (VERSION 1.6.0, kb_schema_version 1.4):
- wiki/decisions/ scaffold — numbered NNNN-slug.md records, own index
(with status vocabulary) and log
- type: decision adds status/decided_on/decided_by/affects/review_on;
supersedes/superseded_by carry history and must be set on both sides
- New ckb-decide skill: records decisions and answers what/why/who/when,
what superseded what, and what is still open. Decision pages are
append-only — a changed mind is a new superseding decision
- Graph gains decided_by and affects edge types
- ckb-ingest routes decisions found in raw material to this format;
ckb-retrieve gains the decisions index as a cascade step
Scriptified the mechanical skills:
- ckb-export-okf/scripts/export_okf.py does the whole OKF transform
(frontmatter remap, link rewriting, index/log regeneration, conformance
validation); --check validates without writing
- ckb-lint/scripts/lint_report.py does the read-only detection half
(conformance, freshness, confidence, retention, decisions, orphans,
graph, index/log, source.yaml); judgment calls stay with the model
Also: removed the duplicate personal quiz skill, fixed stale cbk-quiz
doc paths, gitignored __pycache__.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The last skill clarification (check the shared index store on every
run; fetch if it exists; a write-access user's run creates it there
the first time it's empty) hadn't been reflected in the user-facing
docs, which only described the steady-state case. Also documents the
optional index.ref field, mentioned in the skill schema but missing
from all four docs' examples.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
source.yaml gains an optional index: block declaring where an
already-built index lives (a git repo or a shared resource), so a
user can fetch it instead of scanning the live connector from
scratch. Whether a given user may actually rebuild/publish an index
is now a local, per-user, gitignored source.local.yaml (access:
write|read) that defaults to read-only, letting a team designate one
or two admins per external source instead of everyone redundantly
re-indexing it. ckb-lint's checks against a connector's generated
index now respect the same read/write gate.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
libs/<name>/ now supports a second population mode alongside git-copy
clones: a user-authored source.yaml declares a live external source
(SharePoint, Google Drive, a plain URL, or another connector), and the
new ckb-index-external skill builds a self-contained generated index
for it (index.md/entities/graph/log.md), scoped entirely to that
connector and never blended into the main wiki/.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>