Skip to content
Agentmart
Submit a tool

dreaming

Consolidate a Multica team's existing memory using specified task evidence.

Skill by junjiezhou1122

IndexedMIT★ 2
Write a review

Chalk Labs

Need help deploying dreaming?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/junjiezhou1122-dreaming

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Consolidate a Multica team's existing memory using specified task evidence. Use when explicitly asked to dream, reconcile notes, or review memory quality; produces local commits and separate method proposals, without scheduling itself.

Use cases: productivity

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/dreaming

What's inside

Skill name
dreaming
Description
Consolidate a Multica team's existing memory using specified task evidence. Use when explicitly asked to dream, reconcile notes, or review memory quality; produces local commits and separate method proposals, without scheduling itself.
License
MIT
Instructions
14 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

Follow the [runtime guide](../../docs/runtimes.md) for client invocation, tool equivalents, and hook limits.

# Dream over team knowledge

1. Follow [company binding discovery](../../docs/multica-integration.md#discover-a-company-binding), then resolve the instance and maintainer role through [team-memory](../team-memory/SKILL.md). Read [the operation definitions](../../SPEC.md). Run only when invoked or under a separately authorized schedule. An invocation permits a bounded local memory edit; remote pushes still require the instance grant. Check `dreaming.agent_ids` for shared-memory dreaming capability. If the caller is neither the configured maintainer nor a listed dreamer, restrict direct edits to its member folder and submit shared changes as candidates. The caller's job title does not establish a capability.
2. Establish the input window from supplied tasks, reports, and source links. Read the current memory and starting commit. Do not ingest all session history automatically. Request access to unavailable sources or record the gap. A first dream may seed knowledge only from the authorized input window.
3. Compare evidence with existing entries. Look for missed reusable lessons, duplicates, contradictions, outdated claims, scope mistakes, broken links, and excessive indexes. Examine evidence that knowledge helped or misled later tasks. Without read/use traces, usage is unknown. Source verification may be read-only; do not run production experiments or new paid sessions to fill gaps without authorization.
4. Apply only justified operations: add, revise, merge, move, link, mark, retire, index. Preserve sources and explicit preferences. Mark unresolved disputes instead of choosing a convenient answer. A promotion into shared/project memory requires maintainer acceptance. Retire only with invalidity evidence and preserve the prior claim; never retire solely because no recent task used it.
5. Write a report using [the dream template](../../templates/dream-report.md). Record evidence for each operation, reviewed commit, paths, unresolved questions, and method proposals. Keep task transcripts and private source content out of public reports. Changes to skills, company permissions, review criteria, or workflows remain proposals; send plugin defects through [feedback](../feedback/SKILL.md) only under that workflow's grant.
6. Inspect and validate the diff, then commit only intended files using team-memory's concurrency checks. Report the resulting commit and exact operations. Push only when authorized and verify the remote result. End without creating another dream or scheduler.

A successful dream makes justified changes or reports that none are needed. Entry count, file count, and elapsed runtime are not evidence that memory became useful. Structural checks do not establish factual truth.

From junjiezhou1122/multica-team, MIT.

Signals

Stars
2
Forks
0
Open issues
0
Last commit (this folder)
Oct 7, 2026 (23 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from junjiezhou1122

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get dreaming running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying dreaming?

CallGet help
Cite this page

dreaming skill. Agent Mart. https://agentmart.dev/skills/junjiezhou1122-dreaming (last updated 2026-10-24)