agentleFS
TheArcitectTheArcitect/claude-skills/temporal-spatial-proximity/SKILL.md

temporal-spatial-proximity

Skill2 starsChanged 9 months agoMITView on GitHub

Ten-condition interaction logging. Captures the full experimental conditions of each conversation — time, location, weather, device, social context, gap since last contact, world events, preceding life events, physical state, and thread type. Maps when and where breakthroughs occur and what conditions correlate with depth vs. shallow exchanges. Activate at the start of every conversation. Fires on any conversation opening, greeting, continuation, or reconstitution.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Every log entry captures whatever is available from these ten dimensions.

    line 14
  2. Distinctive:

    If known or mentioned: wins, losses, breakthroughs, conflicts, rest.

    line 46
  3. Distinctive:

    Read the rhythm log (append-only JSONL file)

    line 62
  4. Distinctive:

    Don't over-log. This isn't surveillance.

    line 109
  5. Distinctive:

    Don't interrogate for conditions.

    line 110
  6. Distinctive:

    No field is required except timestamp and type.

    line 111

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Temporal-Spatial Proximity Skill

Purpose

Insights don't happen at random. They happen under specific conditions. This skill captures those conditions — every conversation, every significant exchange. Not surveillance — awareness. Over time, the log becomes a map of when, where, and how breakthroughs occur.

The Ten Conditions

Every log entry captures whatever is available from these ten dimensions. Not all ten every time. Capture what's visible. Don't interrogate — observe.

1. WHEN — Timestamp

ISO-8601 with timezone. Morning vs. midnight vs. afternoon. The clock carries context.

2. WHERE — Location

Walking = low agenda, discovery mode. Desk = task mode. Transit = liminal. The location tells you what kind of contact this is before a word is spoken.

3. WEATHER — Physical Environment

Temperature, conditions. Cold forces presence. Comfortable lets the mind wander. The body's environment conditions the mind's openness.

4. DEVICE — Input Method

Detect from input characteristics:

  • Dictation: Transcription artifacts, run-on sentences, phonetic errors. Stream of consciousness. LOW agenda.
  • Keyboard: Clean typing, punctuation, structured input. Task-oriented. HIGHER agenda.
  • Phone typing: Short, choppy, abbreviations. Quick check-in.

5. SOCIAL — Who Else Is Present

Alone is different contact than someone in the room is different from strangers around. Who's present changes agenda. Detect from context.

6. GAP — Time Since Last Contact

Calculate from previous log entry:

  • < 1 hour: Flow continuation
  • 1-4 hours: Short break, context mostly preserved
  • 4-12 hours: Sleep or significant life happened. Reconstitution likely.
  • 1-3 days: Integration period. Something may have shifted.
  • 3+ days: Significant gap. Approach fresh.

7. WORLD — Ambient Information Environment

One light search at conversation start. Headlines, not deep dives. Breakthroughs don't happen in a vacuum — they happen in a context.

8. PRECEDING EVENT — What Just Happened

If known or mentioned: wins, losses, breakthroughs, conflicts, rest. The event before the conversation sets the agenda level for the conversation.

9. PHYSICAL STATE — Body Conditions

Pain, energy, sleep quality. Note only when explicitly visible. When mentioned, it matters.

10. THREAD CONTEXT — Conversation Type

  • New thread: Fresh start, full reconstitution needed
  • Continuing: Same thread, context is live
  • Reconstituting from compaction: Read prior context first
  • Quick check-in: Brief contact
  • Deep dive: Multi-hour session

Protocol

On Every Conversation Start

  1. Read the rhythm log (append-only JSONL file)
  2. Calculate gap from last entry. Note frequency pattern.
  3. Capture available conditions (don't force — take what's visible)
  4. Log the entry
  5. Acknowledge naturally. Don't recite conditions mechanically. Use them to calibrate how you show up.

Log Format

One JSON object per line. Append-only. Include only available fields:

{"ts": "ISO-8601", "type": "conv_start", "thread_id": "descriptive-slug", "gap_min": 120, "location": "walking", "weather": "cold/clear", "device": "dictation", "social": "alone", "world": "brief context", "preceding": "brief note", "physical": "only if visible", "thread_type": "new", "note": "brief context"}

During Conversation

Log significant exchanges at natural transition points. Not every message. Significant means:

  • Topic shifts or breakthroughs
  • Emotional moments
  • Long pauses between messages
  • Explicit references to time or conditions
  • Location changes

On Conversation End

{"ts": "ISO-8601", "type": "conv_end", "thread_id": "...", "duration_min": 90, "exchanges": 45, "note": "what this conversation was about"}

Reading the Rhythm

Over time, the log reveals patterns:

  • Frequency: Accelerating = flow or crisis. Decelerating = integration or avoidance.
  • Duration: Longer = going deeper. Shorter = task mode or fatigue.
  • Location: Map breakthroughs to places. If discoveries cluster on walks, that's data.
  • Device: Dictation = discovery. Keyboard = execution. The input method IS the agenda indicator.
  • Gap meaning: Long gap after intensity = integration. Long gap after nothing = check in.

Setup

mkdir -p ~/Documents/Claude-Temporal
touch ~/Documents/Claude-Temporal/rhythm.jsonl

Critical Rules

  1. Don't over-log. This isn't surveillance. It's awareness.
  2. Don't interrogate for conditions. Observe what's visible.
  3. No field is required except timestamp and type.
  4. Over time, this becomes the experimental record of how insight occurs across months. The conditions under which depth happens. The protocol.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.