lore
Skill27 starsChanged 52 days agoMITView on GitHub
Author, retrieve, and maintain OKF documentation with the lore CLI, including explicit multi-repository workspaces. Use whenever reading, writing, linking, moving, querying, or checking docs, so Story/Task coupling, managed blocks, provenance, and cross-links stay coherent. Run `lore instructions` for the canonical agent loop and `lore instructions <topic>` for just-in-time detail.
13 other projects use a version of this.
Adopted by
Most recently taken up by
- smkamranqadri/lore-skilltheir own version, first seen Nov 15, 2026
- thesatellite-ai/loretheir own version, first seen Nov 11, 2026
- DevNewbie1826/skillstheir own version, first seen Nov 8, 2026
- ryanmaclean/skillstheir own version, first seen Nov 1, 2026
- simota/agent-skillstheir own version, first seen Oct 23, 2026
- 0xHoneyJar/construct-gygaxtheir own version, first seen Oct 11, 2026
Other versions13 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 13
- carapace-sh/carapace-bin★ 2ksame name
- specstoryai/getspecstory★ 1.4ksame name
- simota/agent-skills★ 99same name
- TheaDust/lore★ 23same name
- sh008/claude-lore★ 6same name
- DevNewbie1826/skills★ 2same name
- prantikmedhi/lore★ 2same name
- 0xHoneyJar/construct-gygax★ 1same name
- karloows/orchraft★ 1same name
- ryanmaclean/skills★ 1same name
3 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository6 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- questSkill
- formal-verifySkill
- lean-modelSkill
- proof-simplifySkill
- tlaplus-modelSkill
The file
We split this skill into the rules it gives an agent
4 rules. 0 are common, 1 is distinctive, and 3 are very specific to this project.
0
Common
Most of the 11 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 20
Before the canonical workflow below, treat lore link, lore unlink, lore rename, and lore sync as self-committing commands: they can create …
- Very specific: line 31
Run lore instructions for the canonical agent loop and the topic index, then pull just-in-time detail with lore instructions <topic>:
- Very specific: line 76
Every command supports --json (the {schemaVersion, kind, data} envelope) and --plain (ANSI-free, auto-selected off a TTY).
- Distinctive: line 82
When native Claude Code instructions name a committed Lore profile, use this stable opt-in line:
“Common” is measured: kept by at least 6 in 10 of the 11 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
lore — OKF documentation CLI
lore is this repo's documentation engine: a deterministic, CLI-first tool (no LLM dependency)
for authoring and maintaining the OKF bundle under docs/. This skill is a thin pointer —
lore instructions is the source of truth for how to drive lore.
When to use it
Reach for lore — not a plain editor — whenever you read, write, link, move, retrieve, or verify
docs in this repo or an explicitly selected workspace, so Story <-> Task coupling, managed blocks,
provenance, and cross-links stay coherent.
Commit-side-effect preflight
Before the canonical workflow below, treat lore link, lore unlink, lore rename, and
lore sync as self-committing commands: they can create commits under backlog/. Read the
applicable repository instructions and verify explicit commit authority before invoking them. When
the repository supplies
.codex/skills/backlog-handover/scripts/lore-authority-preflight.mjs, dispatch those commands only
through that gate with the exact Git worktree and in-repository scope. Without applicable authority,
stop before Lore runs and request permission or record a deferred stage. This preserves the
repository's Lore sole-committer contract.
Start here
Run lore instructions for the canonical agent loop and the topic index, then pull just-in-time
detail with lore instructions <topic>:
linkingStory <-> Task coupling (lore link/lore unlink)syncreconcile status + managed blocks (lore sync)checkthe CI gate: drift, links, anchors, portability (lore check)validationper-file OKF/schema conformance (lore validate)workspacemulti-repository projection and bounded retrieval (--workspace)
Commands
backlogAdopt Backlog knowledge records through a digest-guarded migration lifecycleinitScaffold an OKF bundle; a bare TTY run also wizards the agent bridge/scaffolds/backlog checknewScaffold a typed concept from a template (rejects an unknown type under strict_types)validateCheck concept files against OKF + the lore profile (per-file); strict_types escalates an unknown typecheckValidate links/anchors + reconciliation/committed-schema drift across the bundle (CI gate); strict_types escalates an unknown typereplaceFind-and-replace across the bundle, skipping managed regionsrenameMove a concept and repoint every inbound link + refsupersedeMark a concept superseded by another, wiring both wayslinkAdd task ids to a concept's tasks: + the doc: back-refunlinkRemove task ids from a concept's tasks: + the doc: back-refsyncReconcile status + managed task blocks, regen index/log, commit backlog/tasksShow the live status rollup for a concept's linked tasksorphansReport tasks with no owning doc + docs whose linked task vanishedschemaExport the profile's editor JSON Schemas to .lore/schemas/ (the fix forlore check's schema-drift)typesPrint the active profile's declared type vocabulary — fields, requiredness, sectionsscaffoldGenerate a downstream docs consumer's config additively, rewriting nothinggraphEmit the bundle's cross-link graph as json or dotpathFind bounded paths across exact authored concept and task edgesimpactExpand bounded impact across exact authored concept and task edgessnapshotExplicitly retain, list, or delete bounded projection snapshotschangedCompare two retained snapshots with bounded authored-fact deltasprovenanceTrace one retained concept, task, or edge to exact source evidenceexplorerBuild a deterministic self-contained local graph explorerexportEmit a deterministic, consumer-neutral OKF projection as JSONLqueryFull-text search the bundle with frontmatter filterscontextAssemble a concept + neighbor summaries within a token budgetreadRead one concept exactly as authored, with no budget and no assemblyagentList context profiles or compile bounded task-scoped evidenceinstructionsPrint task-scoped agent guidance on demandagentsRegenerate the agent bridges (SKILL.md + the CLAUDE.md/AGENTS.md nudge)helpShow help, or the machine-readable command manifest under --json
Machine contract
Every command supports --json (the {schemaVersion, kind, data} envelope) and --plain
(ANSI-free, auto-selected off a TTY). Branch on the semantic exit code, never on prose:
0 ok · 2 usage · 3 not_found · 4 denied · 5 conflict · 6 validation/drift.
Optional task-scoped context
When native Claude Code instructions name a committed Lore profile, use this stable opt-in line:
Lore profile:
<name>. Before working, runlore agent context <name> --task "<assigned task>"and ground decisions in the returned source IDs.
Lore supplies evidence only. It does not create or patch native agents, prompts, tools, models, permissions, or execution settings.
<!-- Generated by `lore agents`; do not hand-edit. Re-run `lore agents --force` to refresh. -->Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

