Skip to content
Agent Orchestration
Skill

/daily-okr

Use when planning or closing a daily knowledge-compounding cycle across input, cognition, wiki, behavior, creativity, output, and feedback.

BOOST
From plugin
third-brain-v7-skills
14121 skills4 commands
Install
$ npx -y skills add Mark393295827/third-brain-v7-skills --skill daily-okr --agent claude-code

How it fires

How this skill gets triggered: by you, by Claude, or both.

  • Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
  • You can call itInvoke it directly when you want it.
  • Slash command/daily-okr

Context preview

The summary Claude sees to decide when to auto-load this skill.

Use when planning or closing a daily knowledge-compounding cycle across input, cognition, wiki, behavior, creativity, output, and feedback.

SKILL.md

daily-okr.SKILL.md
name: daily-okr
description: Use when planning or closing a daily knowledge-compounding cycle across input, cognition, wiki, behavior, creativity, output, and feedback.
metadata:
  version: "8.1.0"
  updated: "2026-08-18"
  profile: "stateful"
  assumes: "A daily note or equivalent state record is writable and the user can choose one priority."
  conflicts_with: "Invented completion, seven unrelated priorities, or activity counts without evidence."

Daily OKR

<skill_contract> <input>Date, one priority objective, available time, active project, current inputs, and prior-day evidence.</input> <output>A seven-KR daily cycle linking input, cognition, wiki, behavior, creativity, output, and feedback.</output> <done>Each claimed KR has a durable receipt, and today's feedback changes one explicit next-day action or knowledge state.</done> <non_goals>Seven unrelated priorities, invented completion, activity counting without evidence, or treating scheduled notes as execution.</non_goals>

Run one daily objective through seven linked Key Results. The cycle compounds only when evidence from today changes tomorrow's behavior or knowledge state.

Usage Template

Provide: date, one objective, available time, active project, knowledge inputs, and yesterday's evidence. Optional: a scheduled knowledge-loop note; treat it as input, not proof of execution.

Workflow

<intake>

1. Load the current daily state and any scheduled-loop proposal. 2. Choose one objective tied to an active project or explicit learning goal. 3. Set a total time budget and a minimum viable day. 4. Carry forward only unresolved items with an owner or decision value.

</intake>

<unknowns_gate>

If priority, available time, or completion evidence is missing, return `NEEDS_INPUT` with one narrow probe. If the day is constrained, select the minimum cycle rather than fabricating seven completed results.

</unknowns_gate>

<execute>

Create seven causally linked KRs:

1. **Input:** capture one high-value source or observation. 2. **Cognition:** extract one mechanism, contradiction, or decision implication. 3. **Wiki:** write or improve one durable note with provenance. 4. **Behavior:** execute one minimum behavior linked to the objective. 5. **Creativity:** produce one alternative or falsifiable experiment. 6. **Output:** ship one reviewable artifact or project increment. 7. **Feedback:** record evidence, error, and tomorrow's adjustment.

Each KR needs an artifact or receipt. Keep unfinished work visible; do not convert a scheduled trigger into a success claim.

</execute>

<evaluate>

At close, mark each KR `done`, `partial`, `blocked`, or `skipped` from evidence. Check that at least one input became a durable wiki change and one observation changed a future action. Remove vanity metrics and report the bottleneck.

</evaluate>

<state_contract>

Persist `{run_id, status, attempt, budget, evidence, unknowns, last_error, next_action}` plus date, objective, seven KR statuses, artifact links, bottleneck, and tomorrow adjustment. Append receipts; never rewrite historical completion evidence.

</state_contract>

Failure Protocol

  • `NEEDS_INPUT`: no daily objective or time budget can be inferred safely.
  • `BLOCKED_DEPENDENCY`: a required source, project, or tool is unavailable; complete independent KRs and record the blocker.
  • `VERIFY_FAILED`: a KR has no artifact or receipt; mark it partial or skipped.
  • `BUDGET_STOP`: time is exhausted; close the state and select one next action.

Output Contract

Return `status`, `result` (objective plus seven KR states), `evidence` (artifact links or receipts), `unknowns`, and `next_action` for the next cycle.

Edge Cases

  • Only 15 minutes remain: perform Input -> Cognition -> Wiki as one atomic note, then record feedback; mark other KRs skipped.
  • A scheduled loop note exists but no command ran: record the trigger as proposed work, not execution evidence.

Success Metrics

  • One objective governs all seven KRs.
  • At least one source becomes a verified durable note.
  • The closing feedback changes a named next action.

Quality Gates

  • [ ] Every completed KR has evidence.
  • [ ] Scheduled and executed states are distinct.
  • [ ] Carry-forward work has decision value and an owner.
  • [ ] The minimum day respects the declared budget.

</skill_contract>

Read more
Ships withthird-brain-v7-skills

Verification-first Agent Skills for Codex OS, with optional compatibility adapters for Claude Code, Gemini, Cursor, and Windsurf.

Get the whole plugin

Other skills on third-brain-v7-skills.