event-calendar
Skill36 starsChanged 3 days agoMITView on GitHub
Audit and refresh the desks' scheduled-event calendar (services/intel/data/scheduled_events.json) - elections, jobs/CPI/PCE/GDP, FOMC, OPEC+, IEA/OPEC reports, funding deadlines, Congress, holidays, contract expiries, China data. Run when the desk brief warns "event calendar ends", monthly, or before any election or big political date.
- Reads credentials
1 other project uses a version of this.
Adopted by
Most recently taken up by
- Shubhamkumartrichy/claude-trading-skills-indiatheir own version, first seen Nov 15, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- Shubhamkumartrichy/claude-trading-skills-india★ 64same 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.
In this repository6 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- desk-loopsSkill
- desk-tickSkill
- review-tradesSkill
- desk-tickSkill
- pathiel-agentSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 0 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 13
Thesis: every row becomes an intel catalyst (services/intel/sources/catalysts.py::scheduled_catalysts).
- Very specific: line 35
Every row needs: date, at_utc (or null for all day or unknown), event, kind, coins (oil and/or metals.
- Very specific: line 115
Do not change BEFORE_H/AFTER_H or BLACKOUT_IMPACT without the operator's go.
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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.
Files
event-calendar
Operator, 2026-11-05, after the 11-03 midterms dropped Brent 9% with no election on the calendar: "I NEED YOU TO AUDIT THE EVENT CALENDAR AND TAKE THEM INTO ACCOUNT WHEN YOU MAKE DECISIONS... ELECTIONS, BILLS, ALL THAT".
What uses the calendar:
- Thesis: every row becomes an intel catalyst
(
services/intel/sources/catalysts.py::scheduled_catalysts). - Desk brief: the "Events next 72h" line.
- Blackout: impact-3 rows block new entries
(
services/strategies/event_blackout.py::event_blackout). They are enforced inplace_bracketandtrigger_watch.py. - Generated, not in the file: FOMC decisions, CPI, OPEC+ meetings, EIA,
API, Baker Hughes and CFTC come from
services/intel/calendar_us.py. Update those tables there (FOMC_DECISION_DATES_*, CPI_RELEASE_DATES, OPEC_PLUS_MEETINGS), not in the JSON.
Steps
- Read the file and note
valid_through. Research forward to at least 90 days past today. Use one general-purpose agent with web search, and official sources first:
- US data: bls.gov, bea.gov.
- Central bank: federalreserve.gov.
- Oil: opec.org, iea.org, eia.gov, ice.com, cmegroup.com.
- Politics: congress.gov, state election sites. Also look for scheduled Iran/Hormuz talks, sanctions deadlines and IAEA board dates.
Every row needs: date, at_utc (or null for all day or unknown), event, kind, coins (oil and/or metals; "silver" is the old name of the metals group and still accepted), impact (1 to 3), verified, source. Unverified dates stay
verified: false. Never guess a time.Impact 3 is for events that can move price 2% or more within hours. Those block entries. Keep it to:
- FOMC
- CPI
- OPEC+ ministerials
- national elections or referendums
- a dated war, sanctions or peace deadline
Jobs, PCE, IEA and OPEC monthly reports, funding deadlines and holidays are 2. Elections get their own window:
before_hfrom 24h before the polls close,after_huntil results are in.- Update the file. Bump
refreshedandvalid_through. Keep past rows: they are the record the what-if eval replays. - Check:
.venv/bin/python -m pytest -q services/intel/tests/test_scheduled_events.pypython3 -m services.strategies.evals.event_direction_test(what a short or long through each past event did; information only)
- Commit and push. Say "reconnect pathiel MCP (/mcp) in the desk loops window" so place_bracket and the brief reload the file.
Scan (every 30 minutes)
/event-calendar scan. The desk-loops cron runs it at 17 and 47 past the
hour (operator, 2026-11-13: "every 30 mins"; it started at every 12
hours on 2026-11-05: "a scan every 12 hours to get new events and wire
them into our trading decision"). It adds to the overlay at
$PATHIEL_STATE_DIR/intel/scanned_events.json; it never edits
services/intel/data/scheduled_events.json. A curated row wins on a
(date, kind, title) conflict; the brief shows scanned rows with a (scan)
tag. Keep it cheap: one script, one agent, no more.
- Deterministic half.
python3 scripts/event_scan.py(no LLM): congress.gov committee meetings and hearings for the next 21 days (Foreign Relations, Energy and Natural Resources, Foreign Affairs, Energy and Commerce, Armed Services, Appropriations always; any committee whose title matches Iran, Hormuz, oil, sanctions, OPEC, war powers, NDAA...), new actions on the watched bills (H.R. 1422 and S. 556, the Iran-oil sanctions bills, plus bills on a kept markup's agenda), and Federal Register documents with an effective date or comment deadline in the window. It prints and logs oneEVENT-SCAN ...line. WithoutCONGRESS_API_KEYit saysskipped=congress_...and still runs the Federal Register.--dry-runprints without writing. - Research half. ONE general-purpose research agent with web search,
for the next 21 days from today: dated Iran/Hormuz talks and deadlines,
US-Russia-Ukraine talks, OPEC+ meetings not in
calendar_us.py, Congress floor schedules (cloture votes, NDAA, appropriations, sanctions bills), sanctions deadlines (snapback, waivers, general licenses expiring), summits (G7, G20, NATO, Trump travel), Fed speakers (Powell, Waller, Williams), China data (NBS, trade, PBoC). It returns rows only: date, at_utc (or null), event, kind, coins, impact, verified, source, note. Official page first; otherwiseverified: false. No row without a source. Write each row with
scripts/event_add.py(the one writer; it validates, dedups on date + kind + title, and refuses impact 3):python3 scripts/event_add.py --json '{"date": "2026-11-18", "at_utc": null, "event": "...", "kind": "iran_talks", "coins": ["oil", "silver"], "impact": 2, "verified": false, "source": "https://..."}'Impact 3 only for FOMC, CPI, an OPEC+ ministerial, a national election, or a dated war/sanctions/peace deadline; pass
--allow-impact-3for exactly those and say so in the note. Everything else is 1 or 2. Never--allow-impact-3for a hearing, a report, a speech or a summit. Skip anythingcalendar_us.pygenerates (FOMC decisions, CPI, OPEC+ meetings, EIA, API, Baker Hughes, CFTC).python3 scripts/event_add.py --listshows the overlay;--prune-before <DATE>drops old rows.- End with one line appended to
/tmp/oil-loop/progress.log:EVENT-SCAN-RESEARCH <utc> added=N updated=M refused=R(what the agent found, what event_add took). Say nothing else in the loops window unless a row is impact 3 or a new blackout starts within 24h: then one sentence to the operator. - Promote, do not edit. A scanned row the operator wants permanent
moves into the curated file in a normal refresh (steps above), with a
commit; the scan then reports it as
curatedand stops carrying it.
Rules
- Unscheduled headlines (a deal, a strike) are the thesis's job (news, Polymarket), not this file's.
- A blackout never touches an open position, its stop, target or a close.
- Do not change BEFORE_H/AFTER_H or BLACKOUT_IMPACT without the operator's go.
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.

