claudecodex

peer

Peer skills and orchestrated workflows for Claude and Codex

0 on GitHubMITupdated 5 days agoProductivity
/

Plugin peer from marketplace peer in corrabora/peer-skills-and-workflows. Commands read from the repo's manifest — 4 days ago.

What it installs

Skills (16)

  • bind-agent-workflow

    Establish, refresh or check the operator's local binding of the agent workflow to one repository checkout, inside its ignored .peer-skills-and-workflows directory, by inspecting the repository's guidance, quality-profiles and verification tooling, securing local storage through Git's local exclude f

  • clarify-artifact

    Clarify what a written artifact for people or agents, such as a document, report, plan, instruction set or task brief, must accomplish when unresolved audience, purpose, scope, or intended use would materially change its content. Use before drafting or when revision exposes that ambiguity, including

  • maintain-glossary

    Resolve consequential ambiguity in project vocabulary and maintain concise glossary entries, including during planning; migrate an existing glossary when asked. Reading settled vocabulary or explaining a term alone does not require an edit.

  • peer-decision-making

    Resolve an open decision with one fresh peer and return a jointly supported recommendation to the operator. Use when asked to agree with a peer on a direction, choose between proposals or resolve planning decisions. Ordinary advice, review of existing work and recording an already settled decision h

  • peer-implementation

    Implement a task with one fresh peer using ping-pong pair programming, alternating checks and implementation until its objectives and requirements are satisfied. Use when asked to pair on implementation, from a small script to an entire plan or phase. Recommendation-only and review-only requests hav

  • peer-review

    Review any work with one fresh, read-only companion, debate as equal peers, and resolve agreed changes within existing authority. Use when asked to conduct a peer review of code, research, designs, plans, writing or other work. Does not replace a formal reviewer or adjudicator assignment.

  • plan-coordinator

    Coordinate a repository plan phase through fresh Orca workers (one responsible party, four direct quality reviewers, two peer adjudicators), report each review outcome and stop at the operator gate. Use when asked to run, resume or hand off supervised plan work in Claude or Codex.

  • plan-responsible-party

    Implement one plan phase, one authorized correction batch, or the final landing commit as the implementation's only writer, alone or paired when the operator directs, in a supervised Orca task, returning a candidate with verification evidence. Use when dispatched as the responsible party by a plan c

  • plan-reviewer

    Review one assigned quality set of a plan candidate as a direct Orca reviewer seat, under the target repository's review binding, and report findings without fixing them. Use when dispatched as reviewer R1-R4 by a plan coordinator, in Claude or Codex.

  • record-decision

    Record or revise project decisions as concise ADRs, or migrate existing records when asked. Offer a new ADR only for a costly-to-reverse, surprising choice with a real trade-off. An explicit request to document a decision also selects this skill; routine implementation and reading decisions alone do

  • review-adjudicator

    Independently assess every finding of a plan review wave, then reconcile with an equal peer adjudicator in at most three rounds per finding, recommending without operator authority or edits to the candidate. Use when dispatched as an adjudicator by a plan coordinator, in Claude or Codex.

  • successor

    Enable automatic succession for a task after too many turns or compactions, or hand the current work to a fresh successor now when context is degrading. Preserve decisions, authority and unfinished work across sessions, with clearly named current and retired agents in Orca. Use when asked to enable,

  • write-for-agents

    Write or substantially revise prose that directs another agent's work, including skills, agent instructions, task briefs, handoffs, and operational sections of plans or specifications. Use even when writing is part of a larger task. Excludes code-only changes, mechanical text corrections, and explan

  • write-for-humans

    Write or substantially revise prose for people in documents, reports, explanations, and other artifacts, including documentation produced during a larger coding task. Use for human understanding, decisions, or action. Excludes code-only changes, mechanical text corrections, and instructions solely f

  • write-issue

    Write or substantially revise issue titles and bodies for bugs, proposed changes, and exploratory work, including acceptance criteria. Use when preparing issue text, including within a larger task. Does not perform triage, implementation, or plan decomposition; mechanical text corrections do not nee

  • write-pr

    Write or substantially revise pull request titles and descriptions from the actual change and its verification, accounting for associated issue acceptance criteria. Use when preparing PR text, including within a larger task. Does not perform code review, implementation, or merging; mechanical text c

View source on GitHub

Similar plugins

caveman

claude

Talk like caveman. Cut filler. Keep technical details explicit.

22 skills · 1 command · 3 agents · hooks

JuliusBrussee/caveman 119,462

claude-mem

claudecodex

Persistent memory for Claude Code: the next session starts where the last one left off.

29 skills · 1 MCP server · hooks

thedotmack/claude-mem 116,618

Taste Skill

claudecodex

tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.

1 skill

Leonxlnx/taste-skill 114,391

career-ops

claudecodex

Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications

5 skills · hooks

career-ops-hq/career-ops 82,253

mem0

claudecodex

Cross-session memory and token savings for coding agents.

6 skills · 1 agent · 1 MCP server · hooks

mem0ai/mem0 68,893

pake

claudecodex

Package any website or local web build into a lightweight desktop app with Pake CLI.

1 skill

tw93/Pake 63,463

Maintain peer?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex pluginslisted