Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/MissingPackage/nightshiftWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/missingpackage/nightshift/morning)<a href="https://agentmods.dev/commands/missingpackage/nightshift/morning"><img src="https://agentmods.dev/badge/commands/missingpackage/nightshift/morning/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/missingpackage/nightshift/morning"><img src="https://agentmods.dev/badge/commands/missingpackage/nightshift/morning.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00017 | $0.00346 |
| Opus 5.5 | $0.00007 | $0.00138 |
| Sonnet 5.5 | $0.00003 | $0.00069 |
| Haiku 4.5 | $0.00002 | $0.00035 |
Grade A, and why
morning scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 28d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Assemble the morning report across all projects (runs scheduled via tools/install-schedules.sh; the user reads it from the app over breakfast). Write the report in the user's own language, plain language: every project term explained in half a line or omitted, things cited by title never by ID. Read-only. Order:
- Maps. For every project under ~/Projects with a
HANDOFF.mdchanged in the last 24 hours: ONE line per project — the head line now vs yesterday (objective, value, gap). Numbers from the map and the tail of.harness/tried.md, not reconstructed. - The night. For each project with wake-ups after 23:00 (
.harness/loop-state.jsonand~/.claude/loop-watchdog.log): what was tried and what it said, from tried.md, two lines per project. Loops that died or were revived by the watchdog: say it first. - Decisions. All items in the maps' section "Decisions for the user" (§4), across projects, in ONE batch: numbered, one line each with the default. Never more than six.
- One line of health only if there is something to say: a map over 150 lines, a map whose head line has not moved in two sessions, a silent hook, a timer that did not fire.
Total output ≤ 30 lines. If nothing moved and there are no decisions: three lines and close.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 28d ago Changed · -7 lines · -2 tokens per session e16fd408e75b
- 1mo ago First seen · 29 lines · 19 tokens per session scan A bd7ae6f9591a
morning is a command published in the GitHub repository MissingPackage/nightshift (3 stars, last pushed 29d ago), licensed MIT. It adds 17 tokens to every session and 346 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
project
Manage projects: register repos, inspect, configure per-project settings, and remove.
report
Persist a meaningful worker report for durable delivery to the active project orchestrator.
perf-review-redesign
Replace a broken annual performance ritual with a research-backed continuous + quarterly + annual-comp architecture. Produces redesigned cadence, manager scripts, employee-prep guide, calibration protocol, and a transition plan from old to new. Heavy disclaimer on legal/HRIS implications because changing comp linkage…
talent-psych-safety
Measure psychological safety on a team using Edmondson's 7-question scale (anonymous, longitudinal, team-by-team) plus interpretation rubric and intervention design. Cites Project Aristotle. Refuses "psychological safety training" as primary intervention without leader behavior change. Refuses single-snapshot scoring.…
talent-retention
Retention architecture for a high-performer cohort. Stay-interview script (the practitioner's voice), per-person leverage analysis (1-2 changes that would make this person 5x more likely to stay another 24 months), aggregate pattern across the cohort, system redesign recommendations. Cites stay-interview research…
talent-team-dynamics
Audit a team's dynamics. Hackman criteria check (real team or co-acting group?), Tuckman stage, status-hierarchy map, interdependence map, communication-pattern analysis. Refuses generic "team-building exercises" without diagnostic first. Names which Hackman criterion fails when one does — clarify…