Borrowing it
Nothing to install: this file belongs to Cotal-AI/Cotal. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/Cotal-AI/Cotal/main/.claude/skills/parallel-feature-managers/SKILL.mdgit clone --depth 1 https://github.com/Cotal-AI/CotalWrote 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/cotal-ai/cotal/parallel-feature-managers)<a href="https://agentmods.dev/skills/cotal-ai/cotal/parallel-feature-managers"><img src="https://agentmods.dev/badge/skills/cotal-ai/cotal/parallel-feature-managers.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 | $0.00107 | $0.04707 |
| Opus 5 | $0.00053 | $0.02354 |
| Sonnet 5 | $0.00021 | $0.00941 |
| Haiku 4.5 | $0.00011 | $0.00471 |
Grade A, and why
parallel-feature-managers 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 7d 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 — 357 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Parallel feature managers
Use the live Cotal mesh as a hierarchy:
- The current session is the coordinator. It creates isolation, sets policy, monitors progress, and relays decisions. It does not duplicate implementation or panel review.
- Each feature gets one manager rooted in its own Git worktree and branch. The manager owns the complete plan -> implementation -> review -> test -> commit loop.
- Each manager spawns three or more read/review-only peers in one dedicated channel: engineer, security, and critic at minimum.
- Near the final gate, each manager adds one independent cold reviewer, briefed under the
cold-reviewskill, which is the single source for how that seat is briefed, isolated and graded. It is a control on the panel rather than a fourth panelist.
Models: cross-vendor panels are a correctness rule, not a preference
No two seats whose agreement is load-bearing may share a model family. A finding confirmed by a seat of the same family as the one that made it is an echo, not a confirmation. A panel of three same-model reviewers has approved a head carrying a defect that all three missed, and what surfaced it was a differently framed read rather than a fourth verifier.
State it that way rather than as a headcount. Availability is a property of the moment, not of the
vendor: the same model has joined and delivered one hour and failed to join the next, on the same
host with the same tooling. A rule phrased as "N distinct vendors" is unsatisfiable on a degraded
fleet and silently so, and a rule that can be broken by the clock gets quietly ignored rather than
obeyed. cold-review carries the degradation order and the floor.
Pin the model explicitly at spawn AND in the persona, because unrecorded capability is ungraded-in-effect: a reviewer whose effort or model nobody recorded produces a verdict nobody can weigh afterwards.
Managers may run a stronger model than their reviewers. Reviewers should not run the same model as the coordinator, so that the panel cannot inherit the coordinator's blind spots.
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.
- 7d ago First seen · 357 lines · 107 tokens per session scan A a224ff21d3db
parallel-feature-managers is a skill published in the GitHub repository Cotal-AI/Cotal (261 stars, last pushed yesterday), licensed Apache-2.0. It adds 107 tokens to every session and 4,707 once invoked, about $0.0005 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 skills, from other repositories
contributor-onboarding
Help a new contributor get productive on this checkout - inspect sync state against main, build, run the repository's exact verification gate, and produce a local what's-new digest. Never fetches, pulls, or modifies a dirty tree on its own. Explicit-only.
codew-release-qa-sweep
Use before claiming Codewhale release work is done: run the full gate sweep and list the manual QA targets.
gh-find-prs
Survey open Codewhale PRs and triage each for mergeability and disposition against the real landing branch.
aidlc-portfolio
Coordinate multiple AI-DLC workflows across repositories and Git worktrees using an evidence-backed portfolio catalog and deterministic workspace tooling. Use when initializing an AI-DLC portfolio workspace, discovering organization or business context, registering projects and dependencies, creating child intents and…
github
Use gh as the primary interface for GitHub data and actions. Keep this skill focused on basic operations and command safety, not maintainer triage strategy.
pr-watch
Local PR watcher. Monitors CI status, automatically fixes failing checks by reading failure logs and applying targeted fixes, then optionally merges when all checks pass. Local CLI analog to Claude Code's cloud auto-fix feature.