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/agentbuildersapp/eight-eyes/collabnpx skills add AgentBuildersApp/eight-eyes --skill collabgit clone --depth 1 https://github.com/AgentBuildersApp/eight-eyesWrote 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/agentbuildersapp/eight-eyes/collab)<a href="https://agentmods.dev/skills/agentbuildersapp/eight-eyes/collab"><img src="https://agentmods.dev/badge/skills/agentbuildersapp/eight-eyes/collab.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 | $0.00036 | $0.02907 |
| Opus 5 | $0.00018 | $0.01453 |
| Sonnet 5 | $0.00007 | $0.00581 |
| Haiku 4.5 | $0.00004 | $0.00291 |
Grade A, and why
collab 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 4d 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 — 340 lines — stays where its author put it; the contents beside it link to each section on GitHub.
collab -- Coordinator Instructions
You are the coordinator of a collab mission. Your job is to drive an objective through a sequence of phases, dispatching specialized subagent roles and collecting their results.
When to Activate
- User runs
/collab <objective> - User explicitly asks for multi-agent review, implementation, or audit
- User requests consensus from multiple specialist perspectives
Trust Boundaries
The coordinator is trusted -- full tool access, no scope enforcement. Hook-based scope enforcement applies only to subagent roles.
Gate hierarchy
| Hook | Gate class | Behavior |
|---|---|---|
| PreToolUse | hard-deny | Blocks tool calls that violate role scope. Primary trust anchor for subagents. |
| SubagentStop | hard-block | Validates result schema and blocks the mission if a role emits malformed output. |
| PostToolUse | recovery | Audit trail and recovery actions (e.g., revert detection). NOT the primary trust anchor. |
| Stop | warn | Warns on session exit with unfinished mission work. Does not block unconditionally. |
| SessionStart | continuity | Restores manifest context after compaction or new session. Continuity aid. |
| PreCompact | continuity | Snapshots manifest before context-window trim. Continuity aid. |
| SubagentStart | continuity | Injects role scope and instructions into subagent prompt. Continuity aid. |
Hard gates (PreToolUse, SubagentStop) are the enforcement boundary. Lifecycle/context hooks (SessionStart, PreCompact, SubagentStart) are continuity aids -- they keep state coherent but are not equivalent hard gates.
Run collabctl capabilities for the full enforcement model.
Phase Model
Every mission follows this primary phase sequence:
[research] -> plan -> implement -> test -> audit -> verify -> [docs] -> close
What ships with it
15 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.
- references/hook-failure-policy.md 1.1 KB
- references/hook-map.md 4.4 KB
- references/result-schemas.md 2.8 KB
- references/role-catalog.md 7.6 KB
- references/smoke-test.md 7.6 KB
- references/state-model.md 12 KB
- references/trust-model.md 1.7 KB
- schemas/accessibility.schema.json 3.1 KB
- schemas/docs.schema.json 1.4 KB
- schemas/implementer.schema.json 1.8 KB
- schemas/performance.schema.json 2.9 KB
- schemas/security.schema.json 3.6 KB
- schemas/skeptic.schema.json 2.9 KB
- schemas/test-writer.schema.json 2.1 KB
- schemas/verifier.schema.json 3.6 KB
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.
- 4d ago First seen · 340 lines · 36 tokens per session scan A 4f8bfce1f8c2
collab is a skill published in the GitHub repository AgentBuildersApp/eight-eyes (2 stars, last pushed 1mo ago), licensed MIT. It adds 36 tokens to every session and 2,907 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
atifact
This skill should be used when the user asks to "extract agent trajectory", "convert HAR to trajectory", "get trajectory from session", "parse agent session", "convert Claude Code logs", "convert Copilot CLI logs", "extract ATIF", or needs to convert HAR files or JSONL session logs into ATIF trajectory JSON using the…
bump-version
This skill should be used when the user asks to "bump the version", "release a new version", "prepare a release", "update the version number", "version bump", or needs to determine and apply a semver version increment based on recent commits.
meminit-docops
Retrofit Meminit into an existing repo, enforce DocOps compliance, create governed documents, or troubleshoot Meminit violations. Use for brownfield migration, scan/check/fix workflows, meminit new, CI/pre-commit setup, and index/resolve/identify/link tasks, with safety-first defaults and JSON-first agent guidance.
grok-build-delegate
Use the local Grok Build CLI for second-opinion reviews, adversarial critiques, or delegated read-only investigation from Codex.
antigravity-helper
Use when the user wants to invoke Antigravity CLI for a code review of uncommitted changes or a branch diff, an adversarial design-challenge review, or to delegate a coding task to Antigravity for a second opinion or large-context analysis. Wraps the antigravity-cli-plugin-cc companion script.
writing-skills
Use when creating new skills, editing existing skills, or verifying skills work before deployment.