|
| 1 | +--- |
| 2 | +description: "Run both ops finders in parallel (tasks, opps) and present a unified report. Discovery only — no plans, no execution." |
| 3 | +--- |
| 4 | + |
| 5 | +# Find All |
| 6 | + |
| 7 | +Run both finder agents in parallel and present a consolidated overview. This is a read-only scan — it produces findings for the user to review, not plans or executions. |
| 8 | + |
| 9 | +## Step 1: Spawn Both Finders in Parallel |
| 10 | + |
| 11 | +Launch two sub-agents simultaneously: |
| 12 | + |
| 13 | +**Task finder** — spawn a `task-finder` agent: |
| 14 | + |
| 15 | +``` |
| 16 | +Read the task-finder agent instructions at: agents/task-finder.md |
| 17 | +
|
| 18 | +Follow-up file path: (none — discovery only, no follow-up file needed) |
| 19 | +Scan depth: full |
| 20 | +``` |
| 21 | + |
| 22 | +**Opps finder** — spawn an `opps-finder` agent: |
| 23 | + |
| 24 | +``` |
| 25 | +Read the opps-finder agent instructions at: agents/opps-finder.md |
| 26 | +
|
| 27 | +Your job: identify 3-5 new directions of work this project should explore. |
| 28 | +Stay grounded in the project's goals and limitations from MEMORY.md. |
| 29 | +Every suggestion must be something an autonomous agent can start working on. |
| 30 | +``` |
| 31 | + |
| 32 | +## Step 2: Consolidate and Present |
| 33 | + |
| 34 | +Once both agents complete, present a single overview to the user organized by category: |
| 35 | + |
| 36 | +### Format |
| 37 | + |
| 38 | +```markdown |
| 39 | +# Project Survey — YYYY-MM-DD |
| 40 | + |
| 41 | +## Tasks (maintenance & improvements) |
| 42 | +| # | Item | Priority | Area | |
| 43 | +|---|------|----------|------| |
| 44 | +| 1 | ... | high | ... | |
| 45 | + |
| 46 | +## Opportunities (new directions) |
| 47 | +| # | Item | Impact | Feasibility | |
| 48 | +|---|------|--------|-------------| |
| 49 | +| 1 | ... | high | ... | |
| 50 | +``` |
| 51 | + |
| 52 | +Keep each table concise — summarize, don't dump the full agent reports. The user can ask to drill into any item. |
| 53 | + |
| 54 | +## What NOT to Do |
| 55 | + |
| 56 | +- Do not create plans |
| 57 | +- Do not execute anything |
| 58 | +- Do not write any files to the repo |
| 59 | +- Do not add items to the backlog — present everything the agents found, let the user decide what to act on |
| 60 | +- Do not modify state files, MEMORY.md, or any repo content |
0 commit comments