From co-agents
Conventions for the co-agents per-project memory folder (.coagents/). Use when reading or writing project memory — discussions, decisions, requirements, tasks, research, reviews — so all agents record consistently and nothing is lost between sessions.
How this skill is triggered — by the user, by Claude, or both
Slash command
/co-agents:team-memoryThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Every project the team works on keeps a committed `.coagents/` folder. It is the team's shared,
.coagents/)Every project the team works on keeps a committed .coagents/ folder. It is the team's shared,
version-controlled memory — the trace of what was discussed, decided, required, built, and
reviewed. Read it at the start of a session; write to it whenever a session produces something
worth keeping.
Sibling skills for what to write: [[decision-and-spec]] (specs + ADRs), [[research-method]] (findings), [[facilitation]] (discussion summaries), [[codebase-onboarding]] (the charter on a new repo).
.coagents/
README.md how this memory works (human-facing)
charter.md project principles + stack — READ FIRST every session
discussions/ brainstorm summaries: {YYYY-MM-DD}-{slug}.md
decisions/ ADR-style log: {NNNN}-{slug}.md (one decision per file)
requirements/ specs: {slug}.md
tasks/ task breakdowns: {slug}.md (checklist with owners + status)
research/ sourced findings: {slug}.md (with sources, dates, confidence)
reviews/ review & critique reports: {slug}.md
docs/. Operational/working memory lives in .coagents/. Never keep the same content
in both — link instead.@architect → requirements/, decisions/, tasks/, design docs in docs/@engineer → code, tasks/ (status), decisions/ (implementation decisions)@tester → reviews/ (test plans/verification), test code, tasks/ (status)@reviewer → reviews/; may flag a decisions/ entry (annotate, never rewrite)@researcher → research/, reference material in docs/@scribe → discussions/, decisions/ (recording on the team's behalf), docs/.coagents/ (and your code) under version control so any unintended change
shows up in the diff and can be reverted. That diff is your real guardrail./team-ask able to answer "what does this mean /
why did we decide this" from the record.decisions/{NNNN}-{slug}.md# {NNNN}. {Title}
- **Date:** {YYYY-MM-DD}
- **Status:** Proposed | Accepted | Superseded by {NNNN}
## Context
What's the situation and the forces at play?
## Decision
What we decided, and why this over the alternatives.
## Alternatives considered
- Option A — rejected because…
- Do nothing — …
## Consequences
What this makes easier, harder, or commits us to. Known risks.
discussions/{YYYY-MM-DD}-{slug}.md# {Topic} — {YYYY-MM-DD}
**Question:** one or two sentences.
**Positions**
- Sol (@architect): …
- Max (@engineer): …
- Cass (@reviewer): …
- Ada (@researcher): …
**Tensions:** the real disagreements.
**Decision:** what was agreed (link the decisions/ entry).
**Open questions:** what's unresolved / needs a spike.
tasks/{slug}.md# Tasks: {feature}
Requirement: ../requirements/{slug}.md
- [ ] 1. {task} — owner: @engineer — done when: {condition} — deps: none
- [ ] 2. {task} — owner: @tester — done when: {condition} — deps: 1
Status markers: [ ] todo, [x] done, [~] in progress, [!] needs re-verification.
npx claudepluginhub mohamed-abdelsamei/co-agents --plugin co-agentsProvides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Fetches up-to-date documentation from Context7 for libraries and frameworks like React, Next.js, Prisma. Use for setup questions, API references, and code examples.