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.
npx skills add jason21wc/ai-governance-mcp --skill pair-channelgit clone --depth 1 https://github.com/jason21wc/ai-governance-mcpWrote 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/skills/jason21wc/ai-governance-mcp/pair-channel)<a href="https://agentmods.dev/skills/jason21wc/ai-governance-mcp/pair-channel"><img src="https://agentmods.dev/badge/skills/jason21wc/ai-governance-mcp/pair-channel/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/skills/jason21wc/ai-governance-mcp/pair-channel"><img src="https://agentmods.dev/badge/skills/jason21wc/ai-governance-mcp/pair-channel.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 | $0.00126 | $0.01275 |
| Opus 5.5 | $0.00050 | $0.00510 |
| Sonnet 5.5 | $0.00025 | $0.00255 |
| Haiku 4.5 | $0.00013 | $0.00128 |
Grade A, and why
pair-channel scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Pair Channel
Read procedure.md completely before creating or joining a channel. It carries
the protocol; this file only tells you how to operate it.
The instrument is channel.py, beside this file. Every write to the channel
and every wait goes through it. The protocol's only concurrency guard is
"the session named by TURN: writes; the other waits." Written as a rule it
was voluntary and it failed in another project on the first day; as a script
it is checkable, and its mutating commands run under a lock.
PC=~/.claude/skills/pair-channel
[ -f "$PC/channel.py" ] || PC=~/.codex/skills/pair-channel
python3 "$PC/channel.py" init <channel> --set CHANNEL_NAME=tempcom1 --set REVIEWER_HOST="Claude Code" ...
python3 "$PC/channel.py" state <channel> # TURN / NEXT_SEQ / OPEN / ESCALATED
python3 "$PC/channel.py" append <channel> --role REVIEWER --message msg.md --flip-to IMPLEMENTER --set-open P1
python3 "$PC/channel.py" fill <channel> --role IMPLEMENTER --set IMPLEMENTER_WORKTREE=/abs/path
python3 "$PC/channel.py" ledger <channel> --role IMPLEMENTER --id P1 --decision "…" --commit abc123
python3 "$PC/channel.py" watch <channel> --role REVIEWER # blocks; exit 0 = your turn, 2 = file gone
python3 "$PC/channel.py" check <channel> # every invariant; run when anything looks odd
append refuses: an out-of-turn write, a wrong or duplicate sequence number,
two messages in one file, an unbalanced code fence, and any thread line
beginning with TURN: outside a fenced quote (TURN lives only in STATE). It
stamps the timestamp, updates STATE, and re-verifies. fill (header tables
only) and ledger are the rule-1 carve-outs with the same turn check. Every
mutating command holds <channel>.lock. watch reads STATE, never the thread.
Starting a pairing (REVIEWER does this)
- Both sessions already have their own worktrees (
/start-worktree). The reviewer never writes a tracked file, so its worktree stays at zero commits. - Create the channel OUTSIDE both worktrees in a gitignored place both can
reach — on this framework's repos,
<repo>/.claude/worktrees/<name>.md, which.git/info/excludehides. Confirm withgit statusin every worktree. initwith every placeholder you know; the implementer fills its own rows withfillwhen it holds the turn for message #2.- Write message #1 as a channel test (
procedure.md§0),appendit with--flip-to IMPLEMENTER --set-open P0, thenwatch --role REVIEWERin the background (Claude Code:run_in_background: true). - Give the human ONE prompt for the implementer session that points at the
file (
procedure.md§7). After that the human relays no content — the file carries it all. Establishing the pairing standing-authorizes ordinary channel messages and turn flips; neither role asks again per message. The V-016 cadence still limits review to four boundaries, not every choice.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago First seen · 87 lines · 126 tokens per session scan A 66233b2298f4
pair-channel is a skill published in the GitHub repository jason21wc/ai-governance-mcp (0 stars, last pushed 16d ago), licensed Apache-2.0. It adds 126 tokens to every session and 1,275 once invoked, about $0.0005 per session on Opus 5.5. 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 2026-09-21.
Other skills, from other repositories
autoreview
Structured code review through the shared OpenClaw agent-skills installation.
omh-code-review
This is an OMH code-review workflow skill, projected for Agent Skills hosts (Claude Code, Codex, Cursor, opencode, OpenClaw, pi).
revdiff-plan
Review the last Codex assistant message (plan, analysis, or proposal) with inline annotations in a TUI overlay. Extracts the most recent response from Codex rollout files and opens it in revdiff for review and annotation. Activates on "revdiff-plan", "review plan with revdiff", "annotate plan", "review last response"…
asplos-writing-style
Use when drafting or revising ASPLOS prose — engineering the first two pages that rapid review actually reads, stating the cross-layer insight as one quotable sentence, structuring per-layer mechanism sections, writing bounded claims, and fitting everything into 11 self-contained pages that include figures and…
explore-unknowns
Guide the user through a quadrant walk that maps the unknowns of a task — open by listing the known knowns, then work through known unknowns, unknown knowns, and unknown unknowns one stage at a time, ending with a complete four-quadrant map in the user's hands. Use when a request is ambiguous or underspecified, the…
re0-merge
Review and land an external contribution the way this suite does: gate it against the thesis, land it with the author's credit intact, complete a new skill rather than merging it raw, then approve, credit, and explain before closing. Use when reviewing a pull request, as any collaborator or maintainer, not only the…