mattvideoproductions/MVP-Prompts / skills/foreman
foreman
VerifiedBe the foreman on a big job. One lead model plans the work, hands pieces to helper agents in lanes that never share files, checks everything before accepting it and reports the cost; or runs a long solo build with an owner in the loop. Covers the North Star plan of record, briefs that carry the lead's reasoning, measured gates plus the owner's judgement, strike-based promotion between model tiers, a cost ledger and a mistake register every pass is checked against. Use it when the user types /foreman, says "be the foreman", "be the lead" or "orchestrate this", asks to delegate to subagents or run lanes or passes, or mentions a North Star, three strikes, tiers or promotion between models. Also load it when a task looks too big for one context or needs several agents or days, even if not asked by name; the skill then has you offer it and wait for a yes. Also use it to resume such a project, close a pass, report a session's cost or record a mistake. Not for a change one agent can finish in one sitting.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 32.2 KB |
| assets/NORTH_STAR.template.md | markdown | 4.3 KB |
| assets/brief.template.md | markdown | 1.8 KB |
| assets/config.default.json | config | 2.2 KB |
| assets/discovery-brief.template.md | markdown | 1.8 KB |
| assets/handoff.template.md | markdown | 611 B |
| assets/mistakes.seed.jsonl | other | 21.9 KB |
| assets/prices.json | config | 1.7 KB |
| assets/promoted-brief.addendum.md | markdown | 673 B |
| assets/report.template.md | markdown | 804 B |
| references/audit.md | markdown | 15.9 KB |
| references/briefs-and-handoffs.md | markdown | 10.9 KB |
| references/cost-and-mistakes.md | markdown | 16.5 KB |
| references/field-data.md | markdown | 9.1 KB |
| references/north-star.md | markdown | 11.0 KB |
| references/reference-build.md | markdown | 8.0 KB |
| references/skeleton.md | markdown | 4.3 KB |
| scripts/orch.py | script | 59.5 KB |
assets/NORTH_STAR.template.md
View on GitHub# <PROJECT> - North Star
Lead: <model>. Owner: <name>. Started <date>.
The lead is the only writer. Every agent reads this before working.
Section headings are read by scripts/orch.py. Keep their names, and keep the Status log last.
## One sentence
<the whole project in one line>
## Current state (rewritten every pass, 15 lines at most)
Works: <what a person could use right now>
Fails: <gate> <number> against <bar>
Next: <the single next step>
Unverified by a person: <what no person has yet seen, heard or used>
Waiting on the owner: <what they have been asked to try, and since when>
Spend: <cumulative API-equivalent cost> over <n> passes, cap <N>
## Mandate (owner's words, dated; never reworded)
"<quote>" - <owner>, <date>
What that means here: <lead's reading, marked as the lead's>
## The one thing
<the quality that must be excellent> - owned by <lead>
## Pillars
1. <rule that settles arguments>
## Scope
<what is being built, with counts>
## Decisions and why
- <choice>. Why: <reason>.
## Environment contract
Do not disturb: <what an agent must never interrupt>
Exclusive resources: <each thing only one agent can use at a time (a device, an emulator, the owner's screen, a browser) and how turns are taken; or "none">
Owner's time: <what only the owner can test, how it is batched, and the most they are asked per pass>
Storage: <where large files go, what to clean up>
Irreversible actions: <what needs the owner's yes each time>
Honesty limits: <what must never be claimed>
Privacy: <what never leaves the machine>
### Tool quirks (living; add the moment a trap is found)
- <trap> -> <the form that works>
## Team and working rules
Light: lookups, file surveys, bulk text, mechanical edits. Up to three strikes.
Standard (default): well-scoped tasks that are cheap to check. Promote after one strike unless that pass closed most of the gap.
Senior: judgement across several files, work that is costly to check, and tasks promoted from standard. Needs a reason in the lane ledger.
Lead: the one thing, the task with no clear test yet, and tasks promoted from senior.
Context budget: a worker hands off past <200K> tokens of context. The lead starts a fresh session at each milestone.
## Milestones
M0 <...>
M1 <...> GO / NO-GO
## Gates and bars (bars are set before the work and change only in Bar changes)
| Gate | Command or checklist | Bar | Set by | Set on | Status |
|------|----------------------|-----|--------|--------|--------|
| <name> | <exact command at the shipping configuration, or "owner: checklist <id>"> | <number, or the yes/no evidence that counts> | <owner / lead> | <date> | <PASS n / FAIL n / not run> |
### Known failures (listed so that a new failure stands out)
- <gate>: <number> against <bar>, since <date>. Cause as far as known: <...>
### Bar changes (the old bar stays written here)
- <date> <gate>: <old bar> -> <new bar>. <"owner's words", or "spec change: owner's words"> Why: <reason>
### Owner checklists (what only the owner can run; one session should cover many items)
- <id>: <build or version>. <step> -> <what should happen>. Result: <date, pass/fail, owner's words>
## Lane ledger (one row per lane per tier; a promoted lane closes its row and opens a new one)
| Lane | Task | Tier | Why this tier | Strikes | Gate | Status | Agent |
|------|------|------|---------------|---------|------|--------|-------|
| <name> | <task> | <light/standard/senior/lead> | <required above standard> | 0 | <gate> | <open/running/verify/done/promoted/handed-down/owner/dropped> | <agent ids> |
## Owner notes (every note, in the owner's exact words)
| Id | Date | Owner's words | Routed to | Status |
|----|------|---------------|-----------|--------|
| <N-001> | <date> | <verbatim> | <lane, or "lead"> | <open / done pass n / guessed: what was assumed / asked> |
### Approved and locked (changing any of these needs the owner's yes)
- <date> "<owner's approval, verbatim>": <what is locked, and how the lock is checked>
## Budget
Pass cap: <N>, set by <owner or "lead's proposal, owner not yet asked">, <date>
Cost ceiling: <USD or "none set">
## Status log (append only; dated; real numbers; this stays the last section)
- <date> project founded. Tools and versions found: <...>
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |