evening-recap
Skill9 starsChanged 6 months agoMITView on GitHub
Daily Evening Recap for CEO (21:00 cron). Reports what bot did today, outreach results, pipeline changes, belief updates. Replaces old daily-recap skill. Output = one Telegram message, max 15 lines.
In this repository78 other files it gives its agents.
- anti-pattern-auditSkill
- brand-storytellingSkill
- business-investigationSkill
- ceo-weekly-reviewSkill
- cold-emailSkill
- community-buildingSkill
- company-intelSkill
- competitive-analysisSkill
- competitor-researchSkill
- composio-cliSkill
- conducting-user-interviewsSkill
- content-strategySkill
- copywritingSkill
- daily-recapSkill
- decision-playbookSkill
- designing-growth-loopsSkill
- discovery-debriefSkill
- email-sequenceSkill
- enterprise-salesSkill
- exa-code-searchSkill
- exa-company-researchSkill
- exa-lead-genSkill
- exa-people-researchSkill
- exa-personal-siteSkill
- exa-x-searchSkill
- feedback-trackerSkill
- find-email-by-nameSkill
- focus-keeperSkill
- founder-salesSkill
- fpf-problem-solvingSkill
- github-stars-monitorSkill
- google-calendarSkill
- grafana-monitoringSkill
- humanizerSkill
- hypothesis-trackerSkill
- idea-generatorSkill
- install-openclaw-to-digitaloceanSkill
- install-openclaw-to-hetznerSkill
- install-openclaw-to-ycSkill
- launch-marketingSkill
- lead-enrichmentSkill
- measuring-product-market-fitSkill
- meeting-prepSkill
- metrics-briefingSkill
- morning-briefSkill
- openclaw-guideSkill
- openclaw-user-onboardingSkill
- orthogonalSkill
- outreach-senderSkill
- person-lookupSkill
- ph-launch-timelineSkill
- pipeline-managerSkill
- positioning-messagingSkill
- positioning-workshopSkill
- posthog-analyticsSkill
- prepare-yc-workshopSkill
- pricing-strategySkill
- problem-definitionSkill
- product-hunt-launchSkill
- product-led-salesSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 6 are distinctive, and 2 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.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 9
Read MEMORY.md for: constraint, ICP, current hypotheses.
- Distinctive: line 10
Read today's memory files.
- Distinctive: line 11
Follow output preferences from USER.md.
- Very specific: line 40
If daily_minimum.done == true today — one line of specific process praise (same library as focus-keeper).
- Distinctive: line 48
If calendar has meetings tomorrow → mention prep status.
- Distinctive: line 49
If pipeline has due follow-ups → mention count.
- Distinctive: line 81
If nothing happened today: "Сегодня без GTM-активности." (no streak mention, no shaming)
- Very specific: line 84
Update focus-state.json: daily recap delivered
≈ “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.
Files
[YOUR_PRODUCT] Context
Read MEMORY.md for: constraint, ICP, current hypotheses. Read today's memory files. Follow output preferences from USER.md. For chat platforms: keep under ~15 lines.
Evening Recap
The CEO's second daily touchpoint. Runs at 21:00 via cron. Pure report — no asks, no actions.
Data Sources
memory/autonomous-actions.jsonl— everything bot did todaymemory/outreach-feedback.jsonl— responses to outreachmemory/pipeline.json— pipeline changes todaymemory/focus-state.json— was GTM minimum completed?memory/initiatives.json— initiative progressmemory/hypotheses.json— hypothesis statusmemory/channel-scores.json— which channels performing- Today's
memory/YYYY-MM-DD.md— conversation log
Compose the Recap
Section 1: Today's Actions (3-5 lines)
What happened today (bot actions + CEO actions):
- Outreach sent (and responses if any)
- Content published (with metrics if available: views, reactions)
- Enrichments and research completed
- Pipeline changes
- PR opportunities acted on
- Meetings prepped/completed
Section 1.5: Win (1 line, conditional)
If daily_minimum.done == true today — one line of specific process praise (same library as focus-keeper). If no GTM today — omit entirely (no "0 wins", no shaming).
Section 2: Learning (1-2 lines)
- Belief changes: N (if 0, just say "0")
- Hypothesis updates
- Outreach response rate trends (if enough data)
Section 3: Tomorrow Preview (1-2 lines)
If calendar has meetings tomorrow → mention prep status. If pipeline has due follow-ups → mention count. Otherwise skip.
Section 4: Week Board
{week_board} ({N}/7) | Pipeline: N prospectsWeek board: ●●●○○○○ — one symbol per day Mon→Sun, ● = GTM completed, ○ = not yet. Derive from focus-history.jsonl for current week. Shows accumulation (partial progress visible), not just streak. Never show streak when it's 0.
Output Template
Вечер.
Сегодня:
- [action 1 + result]
- [action 2 + result]
{Win: [specific process praise] — only if GTM done today}
Beliefs changed: N
Завтра: [meeting prep / follow-ups due / nothing special]
●●○○○○○ (2/7) | Pipeline: NRules
- Max 15 lines
- Pure report — no questions, no asks, no approvals
- If nothing happened today: "Сегодня без GTM-активности." (no streak mention, no shaming)
- Neutral tone — no shaming if CEO did nothing, no "0 wins", no streak when it's 0
- Log recap delivery to autonomous-actions.jsonl
- Update focus-state.json: daily recap delivered
Related Skills
- morning-brief — the other daily touchpoint
- daily-recap — legacy skill (evening-recap replaces it)
- structured-log — logging
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.

