dream
Run nightshift dream phase 1 (consolidate) over closed trajectories using the local Qwen model.
Skill by EvolvingAgentsLabs
Chalk Labs
Need help deploying dream?
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.
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/evolvingagentslabs-dream
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.
About
Reviews
Write a reviewNo 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/dream
What's inside
- Skill name
dream- Description
- Run nightshift dream phase 1 (consolidate) over closed trajectories using the local Qwen model. Use when the user asks to consolidate, to run dream, or why nothing reaches candidate.
- License
- Apache-2.0
- Instructions
- 40 lines in SKILL.md · view on GitHub
Agent Skills spec issues
- name: missing
Files (1)
- SKILL.md
SKILL.md (first lines)
# /nightshift:dream
Fase 1 de dream — `consolidate`. **La fase 2 (`verify`) no existe:** es M5 y sigue
prohibida, así que nada llega a `procedure`. `consolidate` **idea siempre**: dibuja el
mecanismo, abstrae desde el dibujo y proyecta síntomas que nadie observó — y lo proyectado
se anuncia como conjetura en cada lugar donde aparece.
Corré el CLI y reportá su salida:
```sh
nightshift dream
```
Si `nightshift` no está en el `PATH`, usá `"${CLAUDE_PLUGIN_ROOT}/bin/nightshift" dream`.
Argumentos útiles, si el usuario los pide:
- `--dry-run` — muestra qué haría sin escribir nada.
- `--lookback-days N` — período a consolidar (7 por defecto).
- `--model "<comando>"` — ejecutable explícito, en vez del backend configurado.
- `--selftest` — el gate de M3-a: corre sobre un set fixture en un store desechable.
Después resumí en dos o tres líneas:
- cuántos grupos se consolidaron y cuántas trayectorias quedaron en `candidate`,
- qué contradicciones se enlazaron (la vieja queda `superseded`, **no** borrada), y
- el caveat honesto: una `candidate` no está verificada. Se inyecta con menos peso y
marcada como no verificada, y así hay que tratarla.
Códigos de salida, porque distinguen tres cosas distintas:
- `0` — consolidó; o no había nada que consolidar; o el modelo dijo que las
trayectorias del período no comparten patrón. Las tres son noches normales.
- `1` — el modelo produjo algo que no se pudo persistir (fuga, ruta en el patrón, JSON
roto). Eso sí es que dream no consolidó, y hay que ir a mirar.
- `2` — no hay modelo disponible. Por defecto dream consolida con Claude Code
(ADR-003) y con `model_backend: "local"` usa Qwen por ollama. Si sale 2, falta el
ejecutable del backend elegido; no es un bug de nightshift.From EvolvingAgentsLabs/nightshift, Apache-2.0.
More from EvolvingAgentsLabs/nightshift
- 0.520score
schedule
EvolvingAgentsLabs
Inspect or set up nightshift's nightly dream run (launchd, systemd user timer, or foreground loop).
SkillBFreshness: BClaude Code +9 - 0.520score
doctor
EvolvingAgentsLabs
Run nightshift's self-diagnosis and end-to-end self-test.
SkillBFreshness: BClaude Code +9 - 0.520score
resolve
EvolvingAgentsLabs
Resolve a conjecture that dream projected — confirmed or refuted, always with evidence.
SkillBFreshness: BClaude Code +9 - 0.520score
why
EvolvingAgentsLabs
Reconstruct the source trajectory behind something nightshift injected.
SkillBFreshness: BClaude Code +9 - 0.520score
sleep
EvolvingAgentsLabs
Run a nightshift sleep cycle on demand — seal the current chapter and consolidate it without ending the session.
SkillBFreshness: BClaude Code +9 - 0.520score
status
EvolvingAgentsLabs
Show what nightshift has captured and what it injected into this session.
SkillBFreshness: BClaude Code +9 - 0.517score
dev
EvolvingAgentsLabs
Working on nightshift itself — the plugin whose code is this repository.
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Aug 27, 2026 (2 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- Apache-2.0
- Repository
- github.com/EvolvingAgentsLabs/nightshift
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from EvolvingAgentsLabs
- 0.520score
schedule
EvolvingAgentsLabs
Inspect or set up nightshift's nightly dream run (launchd, systemd user timer, or foreground loop).
SkillBFreshness: BClaude Code +9 - 0.520score
doctor
EvolvingAgentsLabs
Run nightshift's self-diagnosis and end-to-end self-test.
SkillBFreshness: BClaude Code +9 - 0.520score
resolve
EvolvingAgentsLabs
Resolve a conjecture that dream projected — confirmed or refuted, always with evidence.
SkillBFreshness: BClaude Code +9 - 0.520score
why
EvolvingAgentsLabs
Reconstruct the source trajectory behind something nightshift injected.
SkillBFreshness: BClaude Code +9 - 0.520score
sleep
EvolvingAgentsLabs
Run a nightshift sleep cycle on demand — seal the current chapter and consolidate it without ending the session.
SkillBFreshness: BClaude Code +9 - 0.520score
status
EvolvingAgentsLabs
Show what nightshift has captured and what it injected into this session.
SkillBFreshness: BClaude Code +9
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 dream 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.
Cite this page
dream skill. Agent Mart. https://agentmart.dev/skills/evolvingagentslabs-dream (last updated 2026-10-24)