Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/mphinance/alpha-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/mphinance/alpha-skills/orch-review)<a href="https://agentmods.dev/commands/mphinance/alpha-skills/orch-review"><img src="https://agentmods.dev/badge/commands/mphinance/alpha-skills/orch-review/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/mphinance/alpha-skills/orch-review"><img src="https://agentmods.dev/badge/commands/mphinance/alpha-skills/orch-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00059 | $0.01108 |
| Opus 5 | $0.00030 | $0.00554 |
| Sonnet 5 | $0.00012 | $0.00222 |
| Haiku 4.5 | $0.00006 | $0.00111 |
Grade A, and why
orch-review scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Run a panel review of what's about to be pushed. This is not just code review. It's four different people in the room, each told to look through one lens and nothing else, deliberately set against each other so you get real tension instead of four "looks good"s.
Scope from me (may be empty): $ARGUMENTS
1. Establish the review surface
Figure out what's actually changing, in this order:
- If $ARGUMENTS names a path or draft, that's the surface.
- Else if there are uncommitted changes, use
git status+git diff(and untracked files worth reviewing). - Else use the branch diff vs main:
git diff main...HEAD.
Read enough surrounding context that the reviewers aren't judging a diff in a vacuum. Note whether the surface is code, prose/Substack content, or mixed — that decides seat 2. State the surface in one sentence before fanning out.
2. Fan out 5 reviewers — in a SINGLE message, so they run concurrently
Launch all five as read-only subagents (investigate and report; NEVER edit, write, or commit). Give each the surface and ONLY its lens. The seats are chosen to pull in opposite directions on purpose — do not let them converge into agreement.
Seat 1 — Devil's advocate (claude, read-only). Refute, don't reassure. Build the strongest case AGAINST shipping. What breaks, which assumption is load-bearing and untested, the failure mode nobody's looking at, what bites in two weeks. Default skeptical. Rank concerns by likelihood × damage.
Seat 2 — adaptive, based on the surface:
- If the surface is code → Strategist / ideas (
claude, read-only). What's the bigger swing? The 10x version, the adjacent opportunity, the thing being shipped too small. AND the inverse: is this building something you don't actually need? Concrete, not vibes. - If the surface is prose / Substack content (or mixed) → Editor / your voice (
substack-editor, read-only). Does this sound like Michael, and will a reader get it? Hard-enforce the voice rules — no em dashes anywhere, no markdown tables in Substack content, sign-off is— Michaelonly — then clarity: what's confusing, boring, or buried. Flag every violation with its location. (Voice is a hard rule; for content it outranks the strategy lens.)
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 9d ago First seen · 44 lines · 59 tokens per session scan A f45807d027fc
orch-review is a command published in the GitHub repository mphinance/alpha-skills (21 stars, last pushed 11d ago), licensed MIT. It adds 59 tokens to every session and 1,108 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
sdd-init
Initialize SDD context — detects project stack and bootstraps persistence backend.
review-branch
Review the current branch's diff against base by dispatching atomic-reviewer. No orchestration loop, no spec required — pre-flight before /commit pr or /commit merge.
init
Install the formatters this repository needs, with every command visible before it runs.
merge-conflict-analysis
You are analyzing merge conflicts for PR #${{ pr-number }}.
repo-audit
Audit a codebase (local or remote GitHub/GitLab) against architecture principles and requirements, surfacing drift, risk, and missing decisions.
argos
A command for checking whether an implementation matches its design deliverables. Its Korean description compares the work to the design as part of a completion inspection.