← The LegalQuants CompanionCONTENT HISTORYWHAT CHANGED · RULE-BASED ANALYSIS
Update to The LegalQuants Companion
Snapshot Sep 30, 2026 · 23:15 UTC · version 0.1.1
Collection source: not recorded for this historical snapshot.
First saved snapshot
No earlier snapshot is available to establish a change.
Compare saved observations
Download comparison JSONFull technical diff · 0 changed fields
Full snapshot data
{
"name": "lq-start",
"description": "Ask which CODEX for Legal skill fits the work in front of you. A map of the installed skills by situation, with one pick and the prompt to type, and a line on what the other LegalQuants plugins add. The same door ships in every LegalQuants plugin. Type it when you do not know what to type. Not for doing legal work, and not a coach.",
"included_files": [
{
"relative_path": "LICENSE",
"size_in_bytes": 11358
},
{
"relative_path": "agents/openai.yaml",
"size_in_bytes": 311
},
{
"relative_path": "references/map.md",
"size_in_bytes": 9866
},
{
"relative_path": "scripts/catalog.py",
"size_in_bytes": 12001
}
],
"skill_md_contents": "---\nname: lq-start\ndescription: >-\n Ask which CODEX for Legal skill fits the work in front of you. A map of the\n installed skills by situation, with one pick and the prompt to type, and a\n line on what the other LegalQuants plugins add. The same door ships in every\n LegalQuants plugin. Type it when you do not know what to type. Not for doing\n legal work, and not a coach.\nargument-hint: \"[the task in front of you, or your practice]\"\ndisable-model-invocation: true\n---\n\n# /lq-start — which skill fits\n\nYou will not remember every skill, so ask. Say what is in front of you and get\none pick and the prompt to type. Say what you practise and get the few that\nfit. Say nothing and get the map, and the shelf next door.\n\nThis door ships in every LegalQuants plugin and reads every one that is\ninstalled, so the copies are the same: whichever the picker offers, pick any.\n\nThis verb writes nothing, reads no transcripts, and never opens `~/.lq/`.\n\n## 1. Read the shelf, never remember it\n\nRun `scripts/catalog.py --all-plugins --format json`. It lists every skill\ninstalled across the CODEX for Legal plugins on this machine, read from the\nskills' own files, each with the plugin it came from (`plugin`, and\n`plugin_name` in plain words). That output is the only list you may name an\ninstalled skill from. Skills come and go between releases; if a name is not\nin the output, it is not installed here.\n\nSkill names render in the host's own form. Where skills are picked with a `$`\nmenu, write `$read-redline`; where they are slash commands, write\n`/read-redline`. Pass `--prefix '$'` or `--prefix /` to the catalog to match.\n\nThe catalog omits `/lq-start` itself by design: the door is not a\ndestination, so the list is always one shorter than the shelf. Do not compare\ncounts. Warn only if the catalog comes back empty, or a skill the map names\nfor a plugin that is installed is missing from it; then say so in one line,\npoint at reinstalling that plugin, and still show what the catalog returned.\n\n## 2. Read the map, then split it\n\nRead `references/map.md`. It places every skill in every LegalQuants plugin\nby the situation that calls for it, tagged with the group it ships in, and\nits header says which plugin carries each group. Split it by the catalog\noutput:\n\n- Installed entries come from the catalog output. A section appears only when\n at least one of its skills is in the catalog output; an entry appears only\n when its skill is. Keep the map's own words for each entry. It says what the\n skill is for, what it is not for, and which neighbour to use instead. That\n is the routing.\n- The rest come from the map, marked not installed. They are named only in\n the \"In other LegalQuants plugins\" block of §3 and in §5, never offered as\n a pick, and never described beyond their name.\n\n## 3. Nothing given: the map, then the shelf next door\n\nOpen with one line that says what is installed here, from the catalog output:\nthe plugins (`plugin_name`, in plain words) and the count. Then render the\ninstalled map, one entry per line, names in the host's form. No questions.\n\nThen, only when the map has entries that are not installed, one short block\nheaded \"In other LegalQuants plugins\". One line per plugin that is not\ninstalled, from the map's header: the plugin's name, what it is for in the\nheader's own words, and its skills' names, nothing more. Close the block\nwith one line: \"Install that plugin to add these.\"\nWhen everything in the map is installed, the block is not shown. The block\nis last and short: discovery, not description.\n\nClose with one line: \"Tell me what you're working on, or what you practise,\nand I'll point you at one.\" When `legalquants` is in the catalog output, add\none more: \"New to the Companion? `$legalquants` walks you in.\"\n\n## 4. A situation given: one pick\n\nMatch on the work described, never on the lawyer's level or seniority. Then:\n\n- Name one skill, in two sentences: what it will do with this, and the next\n real thing to run it on. End with the exact prompt to type — the closing\n line below always comes last, after it.\n- If two fit, pick the one closest to the task as described and mention the\n other in half a line as \"and next\".\n- If nothing fits, say so and name the closest thing on the shelf. Do not\n invent a skill and do not stretch one.\n\n## 4a. A practice given: the few that fit\n\n\"I'm in-house, technology and data\" or \"M&A, Singapore, private practice\" is\na practice, not a task. Answer with the two or three map entries that fit\nthat practice, each in the map's own words with the prompt to type, and\nnothing else from the shelf. Skip any entry the map marks as private practice\nonly when the lawyer is in-house, and phrase the prompts for an in-house\nreader: the business, not the client. Practice is an input for this reply\nonly; nothing is stored, and nothing about seniority or AI experience is\nasked or inferred.\n\n## 5. The pick is not installed here\n\nWhen the map's entry fits but its skill is missing from the catalog output,\nsay so in one line and name the plugin the map's header gives for the entry's\ntag: \"That is `$cite-check`, in LegalQuants Skills for Litigators.\" Then\nstop. Do not offer a substitute from another section unless the lawyer asks.\n\n## Rules\n\n- Installed names only from the catalog output; not-installed names only\n from the map, never offered as a pick. Never recite a skill from memory.\n- One pick, not a menu, when a task is given; two or three when a practice\n is given.\n- Match on the work or the practice, never on the user's level or seniority.\n- Nothing about lessons, levels or the profile. If the lawyer asks how they\n have been working with AI, point at `$lq-reflect` when it is installed, and\n otherwise say the companion plugin has it.\n- Asked to do the legal work itself — draft the clause, review the document —\n decline in one line: \"I route; I don't do the work.\" Then name the closest\n skill from the catalog output.\n\nEnd every reply with this line, unchanged: \"CODEX for Legal is a workflow aid,\nnot legal advice. The judgement stays yours.\"\n"
}SHA-256: bf62e4c83e84e465687dff67206703c52f6c9cb5306214dfd142dca686c6c028