sitrep
Skill0 starsChanged 35 days agoMITView on GitHub
Situation report — check every source of truth for this project, report what changed since last check, who holds each open ball, and what the user's highest-leverage moves are. Use when the user says "sitrep", asks to be caught up, or opens a session after a gap.
12 other projects use a version of this.
Adopted by
Most recently taken up by
- silouone/skillstheir own version, first seen Nov 8, 2026
- babs/skillstheir own version, first seen Nov 8, 2026
- hams-ollo/zen-agent-skillstheir own version, first seen Nov 1, 2026
- chrisalehman/bionictheir own version, first seen Nov 1, 2026
- instructa/agent-skillstheir own version, first seen Nov 1, 2026
- EdwardAF-IT/claude-skillstheir own version, first seen Nov 1, 2026
Other versions12 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 12
- instructa/agent-skills★ 139same name
- KrishP147/skills★ 37same name
- skyf0xx/gambit★ 29same name
- iamneilroberts/claude-skills★ 26same name
- jpantsjoha/ai-native-developer-experience★ 13same name
- chrisalehman/bionic★ 9same name
- babs/skills★ 4same name
- silouone/skills★ 4same name
- armenr/modelmux★ 2same name
- hams-ollo/zen-agent-skills★ 2same name
2 more not listed
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
4 rules. 0 are common, 2 are distinctive, and 2 are very specific to this project.
0
Common
Most of the 12 edited versions we read keep these.
2
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 14
Look for a ## Sitrep sources section in the project's CLAUDE.md.
- Very specific: line 18
If no such section exists, infer reasonable sources (recent git log, inbox searches for known counterparties, project tracking files, CI …
- Distinctive: line 30
If nothing changed on a channel, one clause says so.
- Distinctive: line 52
Surface silence as data: when it breaks a pattern; otherwise let quiet channels be quiet.
“Common” is measured: kept by at least 6 in 10 of the 12 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
Sitrep
Produce a situation report: not a status dump, a decision surface. The reader has been away; give them exactly what changed, what's waiting on whom, and what their next move should be — then stop.
1. Find this project's sources of truth
Look for a ## Sitrep sources section in the project's CLAUDE.md. It lists
what to check and how (commands, queries, files, dashboards). Check ALL of
them, in parallel where possible.
If no such section exists, infer reasonable sources (recent git log, inbox
searches for known counterparties, project tracking files, CI status), run
the sitrep from those — and then OFFER to write the ## Sitrep sources
section so the next sitrep is deterministic. Keep the manifest current: when
a new recurring source appears in the project's life, add it.
2. Report structure (in this order)
- Headline — one line: the single most important development, or
"quiet since
<last marker>." - What changed — deltas only. Never re-describe standing state the user already knows; a sitrep that repeats yesterday's sitrep is noise. If nothing changed on a channel, one clause says so.
- Balls in others' courts — who owes what, and how long it's been outstanding. Flag items that have drifted OUT OF THE COUNTERPARTY'S OWN PATTERN (someone who replies same-day going silent for 5 days is signal; a slow replier at 5 days is not). Distinguish "aging normally" from "worth a nudge" and say which.
- The user's court — their open items, ranked by what each unblocks, not by age. Name the single cheapest high-leverage move first. Include anything the user committed to that has not visibly happened (check sent mail / commits / files before assuming).
- Verdict — "nothing on fire" or the opposite, explicitly. If a deadline is approaching, count the days.
3. Conduct rules
- Verify, don't assume. "You said you'd send X" → check whether X was actually sent before listing it as done or undone — and when a draft file exists, read the sent body against it and report any difference. Editors can preview a stale copy; the user may have pasted an old version.
- Deltas need a baseline — keep or find the previous sitrep's state (conversation history, a log file, or the sources themselves) so "no change" is a checked claim, not a guess.
- Surface silence as data when it breaks a pattern; otherwise let quiet channels be quiet.
- End with at most one question or offer — the sitrep's job is to make the user's next 5 minutes obvious, not to open ten threads.
- Blind spots: name any channel you cannot see (e.g., WhatsApp, LinkedIn, a system without access) so the user knows to fill that gap themselves.
- Tables for the courts; prose for everything else; no headers for a quiet day — a short sitrep is a feature.
4. Per-project manifest format (what to put in CLAUDE.md)
## Sitrep sources
- <command to run> # what it reveals
- <inbox/search query> # which counterparties matter
- <file or directory to read> # where commitments/threads are tracked
- Blind spots: <channels the user must check themselves>
- Counterparty patterns: <name — usual response latency>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.

