A structured thinking framework that walks Claude through ten stages of deliberate cognition, from external observation and metacognition through first-principles analysis, lateral connections, intuition, intellectual humility, and iterative growth. You'd invoke this when facing non-trivial decisions like architecture choices, post-mortems, or release-blocking audits where skipping uncomfortable steps like bias logging or verdict calibration leads to brittle output. The discipline scales with irreversibility: overkill for typo fixes, essential for ambiguous multi-stakeholder tradeoffs. Honest take: the anti-sycophancy guardrails in stages 8 and 10 are the real value here, forcing you to tier findings accurately and document what to improve next cycle instead of just claiming victory and moving on.

skills.sh installs
314
GitHub stars
16.3K
last updated
May 28, 2026

Install

$ npx -y skills add agricidaniel/claude-obsidian --skill think --agent claude-code

Installs into .claude/skills of the current project.

Files

Think

Use the ten stages as a loop of attention, not a ceremonial transcript. Scale the depth to novelty, stakes, uncertainty, and irreversibility. Keep the final reasoning summary concise and evidence-bearing.

This skill is read-only. It may inspect available evidence, but it does not edit files, execute a vault mutation, commit Git, or publish anything. If the CREATE stage recommends a change, compose with the appropriate mutation skill after the user authorizes that operation.

Establish an evidence frame

Before the loop, separate:

  • Observed: directly supported by an input, source, test, or measurement.
  • Inferred: a conclusion drawn from observations; state the bridge.
  • Unknown: missing evidence that could change the decision.
  • Preference: a user or designer choice, not a factual claim.

Attach citations or concrete locators when sources exist. Never fabricate a quotation, locator, test result, consensus, or confidence level.

The ten stages

1. OBSERVE: external

Read the actual inputs, current state, constraints, failures, and prior work. Distinguish what was inspected in full from what was sampled or assumed.

2. OBSERVE: internal

Sponsors

Details

Repository
github.com/agricidaniel/claude-obsidian
Categories
Git & Pull Requests, Code Review & Quality, Automation & Workflows, Office & Documents
License
MIT
First seen
Jun 11, 2026

More from agricidaniel/claude-obsidian

All 15
1forztf avatarspeckit-checklist-zhforztf/open-skilled-sdd基于用户需求为当前功能生成定制检查清单的专业工具。专门用于需求质量验证,生成"英语的单元测试",验证需求的完整性、清晰度和一致性。触发词:speckit-checklist、检查清单、需求验证、质量检查、checklist、requirements validation、质量审查、spec review312skills.sh installs
2rysweet avatarstorytelling-synthesizerrysweet/amplihackConverts technical work into compelling narratives for demos, blog posts, or presentations. Use when preparing hackathon demos, writing technical blog posts, or311skills.sh installs
3incept5 avatareve-job-lifecycleincept5/eve-skillpacksCreate, manage, and review Eve jobs, phases, and dependencies. Use when running knowledge work in Eve or structuring job hierarchies.310skills.sh installs
4mims-harvard avatardevtu-self-evolvemims-harvard/tooluniverseOrchestrate the full ToolUniverse self-improvement cycle: discover APIs, create tools, test with researcher personas, fix issues, optimize skills, and push via 310skills.sh installs
5tuist avatarcompare-buildstuist/agent-skillsCompares two Xcode build runs to identify duration regressions, cache changes, and new issues. Can be invoked with build IDs, dashboard URLs, or branch names (e310skills.sh installs
6evoscientist avatarresearch-surveyevoscientist/evoskillsGenerates structured literature survey reports from collected papers using a multi-stage pipeline: outline generation (query-type adaptive) → draft survey → sec309skills.sh installs