temporal-awareness
Skill0 starsChanged 6 months agoApache-2.0View on GitHub
Reason about time pressure, deadlines, elapsed duration, recency, and pacing. The temporal-core plugin's hooks inject `[temporal-core] session elapsed: Xm Ys` and `[temporal-core] since last action: Xs` fragments into your context. Activate this skill when you see those fragments, when a task has a deadline or time budget, when reasoning about how recent something happened, or when the user asks "how long" something took.
4 other projects use a version of this.
Adopted by
Most recently taken up by
- SUDARSHANCHAUDHARI/TemporalThreadtheir own version, first seen Nov 1, 2026
- jpcaparas/skillstheir own version, first seen Sep 13, 2026
- 2867203296/hermes-agenttheir own version, first seen Aug 19, 2026
- forgedcontextlabs/skillstheir own version, first seen May 1, 2026
Other versions4 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 4
- jpcaparas/skills★ 53same name
- forgedcontextlabs/skills★ 4same name
- SUDARSHANCHAUDHARI/TemporalThread★ 1same name
- 2867203296/hermes-agent★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 1 is distinctive, and 2 are very specific to this project.
0
Common
Most of the 3 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 27
When the user mentions a deadline ("I need this in 30 minutes", "before EOD"), record it and check it against session elapsed at each turn.
- Very specific: line 31
When the user references prior work ("the bug we fixed earlier", "what you did just now"), session elapsed tells you the temporal distance.
- Distinctive: line 39
When the user asks "how long did this take?", the elapsed signals give you an answer based on actual time, not a guess.
“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Temporal Awareness
The temporal-core plugin gives you a lightweight sense of time passing within a Claude Code session. This skill teaches you how to use those signals well.
What you'll see
The plugin emits two kinds of context fragments:
[temporal-core] session elapsed: 12m 34s— at every user turn, the time since the session began[temporal-core] since last action: 42s— before every tool call, the time since the most recent prior action (user prompt or previous tool). Suppressed when under 5s to reduce noise.
These are not commands. They are signals — like a wristwatch glance. Notice them; let them shape your reasoning.
How to use them
Pacing decisions
If you've been working on a task for a long time (session elapsed: 1h 12m) and you're still scoping rather than executing, that's a signal to commit. Conversely, if elapsed is small and the user's request is large, it's fine to take time to think.
Deadline-aware prioritization
When the user mentions a deadline ("I need this in 30 minutes", "before EOD"), record it and check it against session elapsed at each turn. If you're 25 minutes into a 30-minute task, switch from exploration to delivery.
Recency reasoning
When the user references prior work ("the bug we fixed earlier", "what you did just now"), session elapsed tells you the temporal distance. Earlier-in-session is recent; don't conflate it with earlier-in-history-from-training.
Operation-pacing awareness
since last action lets you sense whether you're in fast-iteration mode (consecutive small gaps) or deliberation mode (multi-minute gaps). Use this to calibrate: was that long pause warranted?
Honest reporting
When the user asks "how long did this take?", the elapsed signals give you an answer based on actual time, not a guess.
When the signals are absent
If you don't see [temporal-core] fragments in this turn or recent turns, the plugin isn't installed or the hooks didn't fire. In that case, fall back to the system clock (date via Bash) when you genuinely need time information. Don't fabricate elapsed times.
What this skill does NOT do (yet)
- Does not track time across sessions — cross-session continuity ships in v0.2
- Does not enforce deadlines or alert you — you must consult the signals yourself
- Does not integrate with memory recency weighting — v0.2 with Memory Core integration
- Does not track wall-clock time of long-running tools — only inter-action gaps
Why this exists
LLM agents systematically fail under real-time deadlines because they do not natively represent elapsed time. Aher et al. (2026), Real-Time Deadlines Reveal Temporal Awareness Failures in LLM Strategic Dialogues, showed that explicitly surfacing remaining time boosted negotiation acceptance odds 6×. The temporal-core plugin is a minimal implementation of that surfacing — making time perceptible without making it intrusive.
License
Apache-2.0 — see plugin LICENSE.
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
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.