Menu
Search Memex long-term memory. Returns facts, notes, and entities matching the query.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
Deliberately capture what you JUST did as a Memex case — a worked episode (trigger / situation / actions / outcome / lesson) the system derives a reusable procedure from. Use right after finishing a multi-step task, diagnosing a bug, or resolving an incident, or when asked to 'record/log/save how I did this'. Distinct from /remember (auto-routes by shape) and /extract-case (converts existing content) — this is the structured, in-the-moment path.
Resume work from previous /handoff notes. Presents a multi-select list of the most recent handoff summaries, lets the user pick any number of relevant items (or 'load more'), then loads every selected note, summarizes them jointly, and asks what the next steps should be. Use at the start of a fresh session, or when the user says 'pick up where we left off', 'continue', 'what was I working on', 'resume', or 'catch me up on this'.
Tell Memex a surfaced memory was wrong, outdated, or noise. Use whenever you say 'stop suggesting that', 'that's stale / outdated / wrong', 'we removed that', or want a misleading fact to stop surfacing — records a not_helpful outcome on the offending unit(s) and deprioritizes them (the negative-signal loop). NOT for confirming a fix worked (that's a helpful outcome).
Turn a note, a markdown file, or a URL into a Memex case — but only when the content actually contains a reusable how-to / worked episode. Gated: if there's no case in it, files nothing.
Write a technical handoff summary of the current work to Memex — a descriptive, summarizing writeup of what you've been doing, the approach and key decisions, where it stands, and what's next — so you (or another agent) can pick the thread back up later. Use at the end of a workday, when wrapping up an involved conversation, or whenever the user says 'hand this off', 'write up where we are', 'summarize this so I can continue later', or 'I'll come back to this'.
Distill the durable learnings from the current session and save them to Memex, routing each into the right memory layer by shape. Use when the user says 'capture what I learned', 'save the lessons from this session', 'what should we remember from this', or wants the takeaways from the whole conversation kept — as opposed to one named episode (`/case`) or one supplied fact (`/remember`).
| name | recall |
| description | Search Memex long-term memory. Returns facts, notes, and entities matching the query. |
| argument-hint | [search query] |
Query source:
$ARGUMENTS is non-empty, use it as the search query.$ARGUMENTS is empty, look for a "Composed query" block in the additional context — the plugin's UserPromptExpansion hook supplies one when you invoke /recall without arguments. Print a one-line summary of the composed query so the user can see what's being matched.Search strategy:
memex_memory_search/memex_note_search/memex_list_entities scope via vault_ids=["<that vault>"]. If no project vault is set, omit vault_ids (server default). To search every vault, pass vault_ids=["*"]. (Listing without it queries the server's default read scope, which is usually NOT the project vault — the cause of "nothing found" when the briefing shows the vault has content.)memex_procedural_search ONLY (see "Procedure recall" below). Do NOT also run memex_memory_search / memex_note_search for a how-to — the procedural plane is the only home for how-tos.memex_memory_search AND memex_note_search in parallel (both scoped per the rule above). If insufficient, retry with expand_query=true.memex_list_entities. If all fail, say so — do not guess.memex_get_unit_history(unit_id) or memex_memory_search(apply_pre_filter=False).Present results: summarize clearly, include source Note IDs.
Procedure recall: for "how do I X?" queries, the procedural plane is the ONLY source — there is no KV fallback for how-tos.
global → project:<id> → app:claude-code) — answer from there before searching.memex_procedural_search(query="...", kind="procedure") for hybrid BM25+vector search; memex_procedural_get_by_identity(kind="procedure", scope="global", verb="...", context="...") for an exact anchor lookup. The plane carries procedure / strategy kinds under the identity anchor (kind, scope, verb, context); strategies anchor on (scope, verb) with no context.Memory hygiene: when asked about stale facts, call memex_get_lint_flags (omit vault_id to use the active vault, or pass the project vault). Each finding's id is the id field; rule_name discriminates the type. Surface findings for the user to act on — do NOT apply autonomously: a canned resolve --action / memex_lint_apply_winner is server-gated and returns HTTP 403 unless the server runs with MEMEX_LINT_ALLOW_UNATTENDED_APPLY=1. For the full resolve/dismiss workflow use /lint.
Diagnostics: "how is this vault doing?" → memex_get_diagnostics_summary (omit vault_id for the active vault, or pass the project vault) — returns cluster count, top entities, and MW score distribution.
When you invoke /recall with no arguments, a UserPromptExpansion hook reads the conversation transcript, extracts the last MEMEX_CC_RECALL_TURNS turns (default 3, range 1–10), and injects a composed query as additional context. The composed text is enclosed between --- Composed query (last N turns) --- markers so the skill can locate it deterministically. If an explicit query is supplied, the hook does not run — your $ARGUMENTS always wins.