vaibhav0806/startup-skill-pack · Archived

founder-operations

Run evidence-backed founder operating rhythms for priorities, commitments, delegation, meetings, decisions, reviews, and retrospectives.

First seen Aug 19, 2026

Installation

$ npx skills add vaibhav0806/startup-skill-pack --skill founder-operations

Summary

  • Run evidence-backed founder operating rhythms for priorities, commitments, delegation, meetings, decisions, reviews, and retrospectives.
  • Use when a founder needs to choose what matters now, turn work into owned commitments, design or repair a meeting, reconcile operating evidence, or stabilize a repeatable company workflow; route finance, legal, board, policy, risk, continuity, and expansion controls to startup-governance.

Stronger alternatives

This repository is archived — consider an actively maintained alternative.

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from vaibhav0806/startup-skill-pack · top by installs.

npx skills add vaibhav0806/startup-skill-pack

Browse all from vaibhav0806/startup-skill-pack

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Repository health

License LICENSE
Default branch main
Open issues 0
Status Archived

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 5,767 B
  • docs SUMMARY.md 452 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 1 installs

SKILL.md

Founder Operations

Turn current evidence into a small set of owned commitments and review loops. Preserve founder and human authority where trust, people, money, contracts, public statements, sensitive relationships, or irreversibility are involved.

Workflow

  1. Establish the company stage, requested outcome, decision horizon, current priorities, available capacity, dependencies, and authority boundary. Use unknown rather than filling gaps.
  2. Classify each consequential input as observed, inferred, proposed, stale, or conflicting. Record source, as-of date, expected delay, confidence, and permitted audience when relevant.
  3. Read [priorities-commitments.md](references/priorities-commitments.md) for priority selection, exclusions, capacity, WIP, ownership, deadlines, metrics, delegation, and approval gates.
  4. Read [meetings-decisions.md](references/meetings-decisions.md) for the meeting type, pre-read, agenda, decision rights, canonical notes, changed assumptions, action closure, and cancellation test.
  5. Read [reviews-retrospectives.md](references/reviews-retrospectives.md) for weekly/monthly operating reviews, metric freshness, misses, escalation, retrospectives, and recipe stabilization before automation.
  6. Return the current diagnosis, evidence and limits, selected priorities or commitments, finished agenda/log/review artifact, risks and unresolved fields, required human approvals, one accountable owner, deadline or evidence trigger, success metric, and review date.
  7. State the shortest useful handoff when another domain owns the next decision. Use startup-governance for financial, legal, board, policy, risk, continuity, security, privacy, or geographic controls; use startup-hiring for people decisions; use customer-success for post-sale commitments; use the relevant strategy, metrics, sales, fundraising, marketing, or launch skill for domain evidence.

Operating rules

  • Start from the company’s current stage, strategy, constraints, and evidence. Do not import a universal startup cadence, headcount rule, meeting count, runway number, or priority threshold.
  • Maintain an explicit now, next, parked, and excluded view. A request is not a priority until its intended outcome, owner, evidence, and trade-off are visible.
  • Every consequential commitment has one accountable owner, a deadline or evidence-based trigger, a measurable or verifiable outcome, a status, dependencies, and a review date. An owner can delegate work but cannot delegate away accountability without a recorded transfer.
  • Show capacity and WIP limits as a proposed operating constraint for this company, not as a law. When new work enters, record what is displaced, delayed, or stopped.
  • Meetings are a means to an outcome. Cancel or convert a meeting when there is no decision, review, learning, or coordination work that requires the participants; do not preserve a recurring meeting merely because it is on the calendar.
  • Keep a canonical decision record for material decisions: issue, options, evidence, decision owner, decision, rationale, assumptions changed, dissent or uncertainty, effective date, review trigger, and resulting actions.
  • Keep actuals, forecasts, targets, and proposals distinct. Never invent progress, close a missed action silently, or upgrade activity into an outcome.
  • Research, analysis, drafting, logging, and reminders do not authorize sending, publishing, spending, changing a live system, hiring or terminating, signing, changing a contract or price, making a public statement, or committing a sensitive relationship.
  • Founder or other expressly authorized human approval is required immediately before trust-sensitive public or customer claims, employment decisions, money or budget commitments, contracts, fundraising or ownership matters, sensitive relationships, irreversible changes, incident communications, and policy or board actions. The agent may prepare the brief, draft, options, and approval record.
  • Minimize personal, customer, employee, candidate, investor, financial, security, legal, and board data. Use approved restricted systems and the applicable retention/deletion rules; keep reusable skill outputs free of identities, credentials, private URLs, and company-specific values.
  • If evidence is missing, stale, or conflicting, expose the gap, assign a reconciliation owner, and either stop the affected decision or label a narrow provisional recommendation. Do not select the convenient source silently.
  • Automate only after repeated real runs show a stable recipe, explicit owner, observable outcome, exception path, rollback or stop condition, and human approval boundary.

Compact output contract

Use only sections that help the request:

  1. Diagnosis and requested operating outcome.
  2. Evidence table: item | status | source | asof | freshness/expecteddelay | confidence | permitted_surface.
  3. Missing, stale, conflicting, or excluded information.
  4. Recommended priorities, decisions, meeting, review, or retrospective artifact.
  5. Commitment table: commitment | owner | deadlineortrigger | metricorevidence | status | dependency | approval | review_date.
  6. Risks, assumptions, escalation, and required human approvals.
  7. Handoff and the next review date.