Skip to content
Agent Orchestration
Skill

/memory-recall

Check CCC's cross-session memory before building a feature or acting on "earlier work" claims.

BOOST
From plugin
claude-command-center
19721 skills1 command12 hooks
Install
$ npx -y skills add amirfish1/claude-command-center --skill memory-recall --agent claude-code

How it fires

How this skill gets triggered: by you, by Claude, or both.

  • Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
  • You can call itInvoke it directly when you want it.
  • Slash command/memory-recall

Context preview

The summary Claude sees to decide when to auto-load this skill.

Check CCC's cross-session memory before building a feature or acting on "earlier work" claims.

SKILL.md

memory-recall.SKILL.md
name: memory-recall
description: Check CCC's cross-session memory before building a feature or acting on "earlier work" claims.
allowed-tools: Bash

CCC indexes every session's commits, tickets, and transcripts across this machine. Two `ccc` CLI verbs surface that index — use them instead of re-deriving history from scratch or trusting an unverified claim in a prompt.

Before building a feature

Run `ccc shipped "<feature description>"` first. It returns a verdict (`shipped: true/false`), a confidence score, and evidence (repo, commit, ticket) when true.

ccc shipped "csv export for dashboard tables"
  • `shipped: true` with real evidence → don't rebuild it; go look at the

evidence and extend it instead.

  • `shipped: false`, or `true` with thin/no evidence → treat it as unshipped

and proceed.

When a task references earlier work

If a prompt or ticket says "like we did before", "the session that fixed X", or otherwise assumes context you don't have, run `ccc recall "<query>"` to find the session(s) it means.

ccc recall "csv export dashboard"

Each hit carries `session_id`, `title`, `repo`, `date`, and a `snippet` — enough to decide whether to open the full transcript before acting.

Two more verbs: file history and decisions

`ccc history <path>` lists commits (`git log --follow`) and indexed sessions that touched a file, newest first — useful before editing something unfamiliar, to see who touched it and why.

ccc history src/app.py

`ccc decisions "<topic>"` finds sessions whose snippet reads as a decision ("went with", "instead of", ...) rather than just mentioning the topic — a heuristic stand-in until a dedicated decision store lands.

ccc decisions "which queue engine"

All four commands accept `--json` for scripted use and exit non-zero on a missing argument. None is destructive or slow enough to need `run_in_background`.

Out of scope for this skill: no pre-spawn hook wires this in automatically — you decide when to run these, this just tells you they exist and when they're worth reaching for.

Claude Code sessions get a `PostCompact` hook (`hooks/post-compact.py`) that prints a short re-orientation block — ticket ref, last few asks, and this reminder — right after a compaction. Codex has no equivalent event (`~/.codex/hooks.json` only supports UserPromptSubmit, SessionStart, PostToolUse, SubagentStart, SubagentStop, Stop, PreToolUse, PermissionRequest), so a Codex session that just compacted won't get an automatic nudge — run `ccc recall` / `ccc shipped` yourself after a context reset there.

Read more
Ships withclaude-command-center

Your coding agents outgrew your terminal. CCC puts every session on one local board and tells you which one needs you. Not affiliated with Anthropic. Start the next while Claude builds the first. A real share card from one of the maintainer's machines.

Get the whole plugin

Other skills on claude-command-center.