Works with the AI Agents you already use
Ergo
Phased multi-agent build in Herdr: Scout → Architect → parallel specialist coders → blind Reviewer → one squash commit, with an approval gate. Requires Herdr + AiHead CLI.
$5
Ergo
Example session with this skill installed
/ergo Add rate limiting to the public API with per-key limits
- Read your context and instructions
- Compiled the ergo
(Illustrative) Scout context.md -> Architect plan.md with 3 tasks (limiter core / middleware / tests) -> you approve -> 3 coders in parallel worktrees -> blind review NEEDS FIXES (1 major) -> fix loop -> APPROVED -> one squash commit on develop.
Connects securely to your tools. The creator never sees your data.
About this skill
Requires Herdr + the AiHead-Orchestrator CLI (free, MIT: github.com/pank85/AiHead-Orchestrator). Works only inside a Herdr workspace.
Plan it. Approve it. Build it in parallel. Land it as one commit.
Ergo runs one big job as a disciplined pipeline across real agent sessions in your Herdr workspace.
- Scout maps the repo with
file:lineevidence. - Architect writes a plan: disjoint file ownership per task, dependencies, an expert persona per coder and a verification command.
- You approve the plan. No code before that.
- Parallel specialist coders each work in an isolated git worktree on their own files.
- Blind Reviewer judges the combined diff (APPROVED / NEEDS FIXES, scoped fix loop, max 3 cycles).
- One squash commit on develop/main/master.
git reset --hardis forbidden.
Supports --models per-role overrides and mixes different agent harnesses (Antigravity, Pi, Claude Code, OpenCode) in one run.
Why it matters
Focused context per role, fewer overlapping edits, traceable handoffs and an independent review. These are intended benefits, not measured gains; extra agents add cost. Use it for multi-file features and migrations, not one-line fixes.
Requirements
Herdr (herdr.dev) and the AiHead-Orchestrator CLI. See the requirements section and the GitHub repo.
Honest limits
Requires Herdr and the AiHead-Orchestrator CLI; it does not run in a plain terminal or chat-only agent. Not tested end-to-end on every harness. Multiple agents use more tokens than a single-agent run. Product default models are catalog-resolved and need your own access to them (or pin yours with --models).
Requires Herdr and the AiHead-Orchestrator CLI; it does not run in a plain terminal or chat-only agent. Not tested end-to-end on every harness. Multiple agents use more tokens than a single-agent run. Product default models are catalog-resolved and need your own access to them (or pin yours with --models).
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed today
Requires
Permissions
File Scopes
Runs herdr/herdm CLI commands locally. No network calls, no credential access. Never pushes.