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 agentmods add skills/outlinedriven/odin-codex-plugin/ai-collab-protocolsnpx skills add OutlineDriven/odin-codex-plugin --skill ai-collab-protocolsgit clone --depth 1 https://github.com/OutlineDriven/odin-codex-pluginWrote 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/outlinedriven/odin-codex-plugin/ai-collab-protocols)<a href="https://agentmods.dev/skills/outlinedriven/odin-codex-plugin/ai-collab-protocols"><img src="https://agentmods.dev/badge/skills/outlinedriven/odin-codex-plugin/ai-collab-protocols.svg" alt="Measured on agentmods" 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 | $0.00084 | $0.00887 |
| Opus 5 | $0.00042 | $0.00443 |
| Sonnet 5 | $0.00017 | $0.00177 |
| Haiku 4.5 | $0.00008 | $0.00089 |
Grade A, and why
ai-collab-protocols 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 5d 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.
This is a copy
86% identical to ai-collab-protocols — 63 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Spot the protocol gap, name the better tactic, point at the durable handle. Small surgical interventions, not a lecture.
Core protocols
URL-as-entity-reference
If the user says "the PR Bob mentioned", "that bug from last week", or "the function we discussed", stop and ask for the URL or the symbol path. Why: names are ambiguous in long-context sessions and unrecoverable across sessions. A stable URL — GitHub PR comment permalink, MCP resource URI like @github:pr/owner/repo/123#comment-456, file:line reference — survives compaction and enables exact match. The chat that prompted this skill explicitly named this as the highest-signal collaboration tip.
Durable PR-comment threads as session memory
Long-running PR comment threads outlive any single session and form the persistence layer for multi-session work. Prefer leaving a comment on the PR over a chat-only handoff. Why: the next session — yours, a colleague's, or a future agent's — can resume from the thread without replaying context. Chat is ephemeral; PR comments are addressable.
Fit the protocol
When a project has an AGENTS.md, CLAUDE.md, or .clinerules, read it before acting. When the project has none and the work is non-trivial, propose authoring one — defer to init for AGENTS.md authoring rather than re-doing it here. Why: project-level rule files are the cross-tool agent-config convention; fighting them creates drift across sessions.
Anti-patterns to flag
- Screenshot-only context — loses URL grounding, copy-paste, and search. Pair every screenshot with the URL or text export.
- Unanchored pronouns — "the PR" / "that function" / "the bug" hallucinate badly in long contexts. Demand a URL or symbol path before continuing.
- Token-usage / LOC framing as a quality proxy — quantity is not capability. The chat that prompted this skill explicitly rejects this framing; surface the rejection if the user reaches for it.
Optional 2026 research extensions (verification status)
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.
- 5d ago First seen · 51 lines · 84 tokens per session scan A fc3105b0214e
ai-collab-protocols is a skill published in the GitHub repository OutlineDriven/odin-codex-plugin (15 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 84 tokens to every session and 887 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. It is 86% identical to ai-collab-protocols, differing in 63 lines, and is treated as a copy.
Other skills, from other repositories
duet
Two-party posture — user as director, agent as executor; every fork, tradeoff, or choice surfaced via batched AskUserQuestion with a recommended default. Use when the user invokes /duet, says "ask before" / "pair with me" / "human-in-the-loop", or for aesthetic/architectural/irreversible decisions.
ai-collab-protocols
Surface in-task-collaboration protocols when the user describes an AI workflow informally — URL-as-entity-reference, PR-comment threads as session memory. Trigger when the user names entities by colloquial label instead of stable URL, asks "how should I structure this for Claude", or describes a multi-step Claude…
tidy
Tidy-operations dispatcher. Invoke on "tidy", "clean up", "tidy this file/memory/workspace/git/docs", or when active context (current file, diff, stack, memory directory) has structural rot to resolve before touching behavior. Detects target domain from context and routes to the sibling skill. Requires explicit target…
inbox
Drain parked decisions — read a tagged digest of every blockedonquestion task's open questions, pick which to answer now, deep-dive each via AskUserQuestion, and record the answer through the answer verb. One pass, suggest-don't-force, no self-loop.
setup
Use when superb's skills need their dependencies installed or checked — "set up superb", "install the dependencies", "why does pipeline say superpowers is missing", after a fresh plugin install, or on a new machine. Also use when a skill fails complaining that a superpowers sub-skill cannot be found.
ghostty
Control Ghostty terminal emulator via CLI. Use when managing windows, tabs, splits, fonts, or configuration for Ghostty.