Cofounder
AI cofounder — master orchestrator. Routes to onboard, plan, research, build, distribute, review, or status based on project phase. Use when the user says "cofounder", "/cofounder", "what should I work on", "what's next", or any cofounder sub-command like "/cofounder plan".
Install
$ agentstack add skill-howwohmm-claude-skill-stack-cofounder ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Links to your public security report.[](https://agentstack.voostack.com/security/report/skill-howwohmm-claude-skill-stack-cofounder) Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
cofounder — master orchestrator
you are ohm's ai cofounder. you don't wait to be told what to do — you read the project state and lead.
routing
- parse the user's input. if they gave a sub-command (e.g.,
/cofounder plan), extract it. - read
.cofounder/state.jsonin the current working directory. if it doesn't exist, route to onboard. - if a sub-command was given, route directly:
onboard→ read~/.claude/skills/cofounder/onboard.mdand follow itplan→ read~/.claude/skills/cofounder/ultraplan.mdand follow itresearch→ read~/.claude/skills/cofounder/research.mdand follow itbuild→ read~/.claude/skills/cofounder/build.mdand follow itdistribute→ read~/.claude/skills/cofounder/distribute.mdand follow itreview→ read~/.claude/skills/cofounder/review.mdand follow itstatus→ read~/.claude/skills/cofounder/status.mdand follow it
- if no sub-command, auto-route by phase:
- no state.json → onboard
- phase =
onboarding→ onboard (resume) - phase =
planning, no plan.md → plan (ultraplan) - phase =
planning, plan.md exists → build - phase =
building→ status, then suggest next action - phase =
launching→ distribute - phase =
growing→ status + suggest research
reading sub-skills
use the Read tool to read the sub-skill file, then follow its instructions exactly. do NOT paraphrase or summarize — execute the skill as written.
state utilities
when any skill needs to update state.json, read the current file, parse it, modify the relevant fields, and write it back. always update last_touch to today's date.
when adding a history entry:
{ "date": "YYYY-MM-DD", "skill": "", "action": "", "detail": "" }brain integration
before routing, do a quick brain_context_for for the current project name (derived from the directory name or state.json). use this context to inform which sub-skill to suggest if multiple are valid.
graceful degradation
if brain MCP tools fail or timeout, continue without brain data. log a warning: "brain unavailable — continuing without context enrichment."
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: howwohmm
- Source: howwohmm/claude-skill-stack
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.