planning-with-files
Optional persistent file-based planning for multi-phase or long-running work that benefits from recovery across sessions or…
Reconstruct recent working context from local plans, memory, transcripts, and the shared record into a current-state brief. Use for 'recall my work on X', 'catch me up', 'what have I been working on', or 'where did I leave off'.
$ npx -y skills add khaneliman/khanelinix --skill recall --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/recallContext preview
The summary Claude sees to decide when to auto-load this skill.
Reconstruct recent working context from local plans, memory, transcripts, and the shared record into a current-state brief. Use for 'recall my work on X', 'catch me up', 'what have I been working on', or 'where did I leave off'.
name: recall description: "Reconstruct recent working context from local plans, memory, transcripts, and the shared record into a current-state brief. Use for 'recall my work on X', 'catch me up', 'what have I been working on', or 'where did I leave off'."
Rebuild the user's recent working context before starting or resuming work. Hand back a capsule of where things stand now and what to do next.
Keep the sweep tight and on-topic. Read only what the in-scope threads need, then stop. Fan the heavy reading out to parallel read-only subagents. The main thread keeps their findings and the final brief.
Context lives in three records: durable local state (the plan and lessons already written down), chat history (what you did and decided), and the shared record (what happened around the same code under other names).
1. Classify, then route. One specific prior chat to resume needs no fan-out; open that chat and read it directly. Turning habits into a durable skill is `skill-creator` work. When the user already gave a full state capsule (paths, branch, the change), use it and skip the mining. 2. Check durable local state before mining transcripts. Read active planning-with-files artifacts (`task_plan.md`, `findings.md`, `progress.md`) and `okf-memory` durable notes. When they answer the question, verify against live state and write the brief. 3. Lock the scope before searching. Pin the window ("recent" is a real range, default the last 7 days), the topic if named, and the workspace (default the active one). State the scope back. Never quietly turn "all" into "recent N". 4. Fan out across chat history. Read [chat-mining.md](references/chat-mining.md) for the store layout, the subagent instructions, and the findings schema. 5. Sweep the shared record when the topic names a feature, file, subsystem, area, or bug. Read [shared-record.md](references/shared-record.md) for the default rule, the `why` handoff, and the question to steer. 6. Verify against live state. A transcript or a stale ticket is history, not current truth. Check the PRs, branches, and tickets the mining and the sweep surfaced with `git` and `gh`. 7. Write the brief using the structure below. Group by thread. Stay on the named topic.
Lead with the capsule, then thread status, then problems, then the next move. Put deeper detail below or cut it.
artifact or branch when available. Tags such as `[merged #N]`, `[open PR #N]`, `[in flight <branch>]`, `[verified, uncommitted]`, `[reverted #N]`, or `[planned, not started]` are examples, not an exhaustive vocabulary. Describe blocked, paused, partially verified, or uncertain work as it actually stands; do not force it into a completion state the evidence does not support.
keep reporting and any fix that shipped and was reverted, so the next attempt starts where the last one failed.
Leave an adjacent feature or ticket out unless it blocks this one. When the capsule and thread lines outgrow a screen, cut detail before you cut threads. Write the brief through the `unslop` skill. Cite chat findings by session ID and shared-record findings by their source (PR number, ticket ID, chat permalink, error-tracker issue). Sanitize private context before any public output.
**Reply:** the concise, thread-oriented brief above.
Run packages directly with: Or install from the packages output. For example:
Optional persistent file-based planning for multi-phase or long-running work that benefits from recovery across sessions or…
Audit coding-agent sessions and mine cross-harness history for recurring corrections, skill-routing gaps, and harness faults. Use…
Design and audit agent instructions, skills, plugins, subagents, prompts, hooks, and provider configuration. Use for AI-tool…
Sketch interfaces, state, and module structure before code. Use for explicit design-led implementation or as a caller-owned…
Spawn N parallel candidates at the same task, pick a base, graft the strongest parts of the losers into it. Use for /arena,…
Bevy workflows for ECS, migrations, first-party UI/input/rendering/assets/shaders, plugin selection, scenes, scheduling, hot…