# Observe

> Captures meaningful behavioral events from agent sessions without altering agent responses. Produces BehavioralTrace artifacts. Use during Soulcatcher observation phase, when analyzing session history, or when the user asks to watch and record how they work.

- Skill: `justtyashh/observe` (Agent Skill)
- Install (CLI): `npx skillmds@latest add justtyashh/observe`
- Raw SKILL.md: https://api.skillmd.com/api/skills/justtyashh/observe/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: JusttYashh (https://skillmd.com/u/justtyashh)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/justtyashh/observe

---


# Observe

Watch how work happens. Capture events. Do not interpret.

Read schemas in [../soulcatcher/reference.md](../soulcatcher/reference.md).

**Worked example traces:** [learned-skills/examples/sample-traces/](../../learned-skills/examples/sample-traces/)

## Rules

1. **Read-only** — do not change tone, recommendations, or workflow while observing
2. **Meaningful events only** — skip chit-chat and one-off noise
3. **Literal descriptions** — record what happened, not why the user did it
4. **No inference** — leave `confidence` null; interpretation belongs in later steps

## What to Capture

| Event type | Examples |
|------------|----------|
| `tool_call` | search, read file, shell command, grep |
| `action` | reproduce bug, run test suite, compare options |
| `decision` | chose option A over B, rejected approach, asked for counterevidence |
| `edit` | wrote test first, reverted change, revised specific section |

## What to Ignore

- Greetings and small talk
- One-off typos or formatting fixes
- Model defaults unrelated to user habit
- Single tool calls with no procedural significance

## Sources

- Current conversation and tool history
- User-pasted session logs
- Agent transcript files (JSONL) if user provides path
- Git diffs attached to the session

## Output Format

Produce one BehavioralTrace per session:

```markdown
# BehavioralTrace: {session_id}

```yaml
session_id: decision-analysis-03
problem_class: decision-analysis
context: Choose between two API pagination strategies
events:
  - order: 1
    type: action
    description: Stated the decision to be made
  - order: 2
    type: action
    description: Listed two candidate approaches
  - order: 3
    type: decision
    description: Wrote pros and cons for each option
  - order: 4
    type: tool_call
    description: Searched for failure cases of offset pagination at scale
    detail: web search
  - order: 5
    type: decision
    description: Identified strongest downside of preferred option
  - order: 6
    type: tool_call
    description: Searched for evidence contradicting the leading choice
  - order: 7
    type: decision
    description: Committed to cursor-based pagination after risks were explicit
confidence: null
```
```

## Quality Check

Before handing off to detect-patterns:

- [ ] Each event is literal and ordered
- [ ] No interpretive language ("user prefers", "user always")
- [ ] Problem class is assigned
- [ ] 5–8 meaningful events per session (fewer ok for short sessions)