/analyze-processes

name: analyze-processes
allowed-tools: mcp__memorylane__browse_timeline, mcp__memorylane__search_context, mcp__memorylane__get_activity_details, mcp__memorylane__get_user_context, mcp__memorylane__list_patterns, mcp__memorylane__get_pattern_details, mcp__memorylane__search_patterns, mcp__memorylane__get_db_path, Read, Grep, Write(./process-analysis*.md), Task, Bash(sqlite3 -readonly -safe *), Bash(cp * /tmp/memorylane-analysis.db), AskUserQuestion, mcp__plugin_memorylane_memorylane__browse_timeline, mcp__plugin_memorylane_memorylane__search_activity, mcp__plugin_memorylane_memorylane__get_activity_details, mcp__plugin_memorylane_memorylane__get_work_profile, mcp__plugin_memorylane_memorylane__list_patterns, mcp__plugin_memorylane_memorylane__get_pattern_details, mcp__plugin_memorylane_memorylane__list_organizations, mcp__plugin_memorylane_memorylane__list_teams, mcp__plugin_memorylane_memorylane__list_devices
description: >
  Find a person's repeated tasks from their screen activity, and what each one is
  worth automating, with the time and money saved. Outputs the numbers and data a report
  is built from, not the visual. Use to analyze processes, mine workflows, or see what is
  automatable and what it saves. Every figure is labelled and grounded, never made up.

Analyze Processes

Turns raw screen-activity rows into a defensible breakdown of a person's repeated work, what it costs, and what to automate. A separate packaging skill renders the visual. **Your job is truth and numbers, not persuasion.** Bar = **80/20**: directionally accurate, grounded, never fabricated. Can't ground a claim? Weaken its label or drop it.

Before you start

Read `${CLAUDE_PLUGIN_ROOT}/references/data-sources.md`. It says whether to use the local or remote tools, and how the remote ones differ. Tool names below are the local ones.

Tools can load on demand, so search for them first. If neither the local tools (`search_context`) nor the remote tools (`search_activity`) turn up, stop and reply with exactly:

> MemoryLane isn't connected yet. In Claude, open **Settings > Connectors** and sign in to **MemoryLane**. If you use the MemoryLane desktop app, open **Settings > Integrations** there and click **Connect** or **Reconnect** next to **Claude Cowork**. Then run this again.

Do not suggest plugin settings, Retry, or toggling the plugin.

Data limits (single source of truth for basis labels)

Verified against the MemoryLane MCP + storage code; everything below references these. These describe the local tools; `${CLAUDE_PLUGIN_ROOT}/references/data-sources.md` lists what differs on remote (real end times, exact totals, raw app names, no scrubbing):

• **Duration:** bulk tools (`browse_timeline`/`search_context`) return **start only**; end is stripped. Only `get_activity_details` gives a real `[start -> end]`. Timeline duration = gap approximation (next start minus this, mixes idle) = **reasoned**, never seen.
• **Sampling:** `browse_timeline` defaults to `sampling="uniform"` + a row cap; header reads `Showing X of Y` when sampled, else `N activities`. Sampled counts = lower bound. `search_context` no-query = `recent_first` (drops oldest). Never count/sum from a sampled pull.
• **No aggregate tool:** totals are reconstructed by enumerating rows (inheriting the sampling + duration limits).
• **No case ID:** instances are reconstructed. IDs in `window_title` (PR/ticket #, doc) = **seen**; invoice/order IDs live in OCR (`get_activity_details`), sparse = **reasoned**.
• **No interaction data:** clicks/keystrokes never stored, so pain (re-keying) = **reasoned** from summaries/OCR.
• **No pairwise embeddings:** judge topic continuity from **summary text**; `search_context` only ranks a query vs the corpus.
• **Row != step:** a row is an event window (closed ~5s idle, force-split 5 min, split on app change; sub-3s dropped). One task spans many rows.
• **Site names:** the timeline app name is already the site for browser rows. OCR is only in `get_activity_details`.
• **One person per pull:** a local DB or a remote device.
• _Non-blocking code follow-ups that turn reasoned into seen:_ add `endTimestamp`+`durationMs` to the timeline and a `get_activity_stats` aggregate tool (primitives already exist).

0. Operating Principles

1. **Three bases on every claim.** **seen** = a raw unsampled count or directly-read value. **reasoned** = inferred from clear signals (segmentation, gap-durations, instance grouping, topic). **estimated** = filled from role/industry/location (rate, fraction, annualization). **Weakest-link:** any aggregate with an estimated input is **estimated**. 2. **Evidence trail.** Every seen/reasoned claim cites activity/session IDs; no evidence = estimated at best. 3. **Scope = n=1.** One observed person; tag aggregates "single observed user (n=1)". No team/FTE-for-team total unless headcount is supplied (then estimated, show multiplier, mark unverified). Print "Based on 1 observed person" in meta. When you do extrapolate to a headcount, keep the measured (observed-people) figure as the lead/anchor and mark the company-wide figure illustrative/directional/lower-range, never the headline. Note: an n=1 report scoped to a role/department is structurally identifying to that client's leadership even with no name printed; anonymization here is a floor, not a guarantee, so this is a consent/scoping conversation with the client, not something the skill's wording alone can fully solve. 4. **Ranges, not false precision.** Every headline/annualized figure ships low/base/high with the dominant uncertainty named. 5. **Reasoning log is first-class.** Per candidate, record why it is/isn't a process (common-path coverage + case-n). 6. **Review inside every stage** (gate re-derives that stage's numbers + checks labels before moving on). 7. **No fabricated granularity.** Activity-level steps only; never click-by-click. 8. **Automation grounded in real tools** (from context); else Cowork (judgment/generatio