New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Nippo Daily Reports

by nwiizo in nwiizo/nippo

Turns recent Claude Code, Codex or opencode work into Japanese daily reports, reflections, reviews and trend summaries saved in a reports folder.

MITAuto-check passedProductivity & Automation

Install Nippo Daily Reports

skills CLI
$ npx skills add nwiizo/nippo --skill nippo -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install nwiizo/nippo nippo --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/nwiizo/nippo.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/nippo .claude/skills/nippo && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
nippo
GitHub stars
180
Token cost
~2.3k tokens
SKILL.md length
1,110 words
Files
2
Skills in repo
2
Repo updated
First seen
Licence
MIT

At a glance

Turns recent Claude Code, Codex or opencode work into Japanese daily reports, reflections, reviews and trend summaries saved in a reports folder.

  • Works in 12 steps: If the current workspace is this… → Otherwise prefer nippo collect ... when… → If neither is available, stop and tell… → …
  • Writing a daily report from what the agent did today
  • SKILL.md covers Inputs, Workflow, Mode Defaults and Rules, plus 1 more section
  • Calls cargo and gh

What it does

The skill reads recent agent work and writes a report under `reports/`. A mode argument selects the kind: daily, brief, reflection, guide, report, review, insight, trend, plan or ledger. Optional arguments set the number of days, a project filter, a source of claude, codex, opencode or all, and a period such as today, yesterday, this week, last month or the one before. Examples include `$nippo daily codex` and `$nippo review 90 codex`.

The agent runs `nippo collect` through the installed CLI, or through `cargo run` when inside the nippo repository so it uses the checked-out code, and stops to ask you to install `nippo` if neither exists. The source defaults to auto, and brief mode saves the summary directly. Ledger mode skips collection: `nippo ledger` gathers the Unclear points sections of daily reports into `reports/ledger.yaml` and prints a verdict that the agent relays as is, while `nippo ledger --export` writes candidate fix rules that a person must curate before anything goes into CLAUDE.md or AGENTS.md. The reports are in Japanese.

When your agent uses it

  • Writing a daily report from what the agent did today
  • Looking back over a week or month of agent work for patterns
  • Summarizing work for one project from a single agent's logs

Example prompts

  • “$nippo daily all”
  • “Write a weekly review from my Codex work logs for last week.”
  • “Make a reflection report of what I did in the api-gateway project yesterday.”

Requirements

  • The `nippo` CLI, or Cargo inside the nippo repository
  • Work logs from Claude Code, Codex or opencode on the same machine

Workflow steps

12 steps, taken from the first numbered list in SKILL.md.

  1. If the current workspace is this repository, prefer cargo run -q -p nippo -- collect ... so the skill uses the checked-out implementation…
  2. Otherwise prefer nippo collect ... when nippo is already installed.
  3. If neither is available, stop and tell the user to install nippo.
  4. Treat daily as an alias of the default daily report mode.
  5. Default to --source auto. Override when the user explicitly asks for claude, codex, opencode, or all. The collector removes deterministic…
  6. For brief, save the summary output directly and stop.
  7. For ledger, do NOT call nippo collect. Run nippo ledger (or cargo run -q -p nippo -- ledger) which scans reports/nippo-*.md (daily…
  8. For plan, do NOT call nippo collect. Find the newest reports/nippo-*.md by the date in its filename and read it. Also read…
  9. For other modes, read the collected JSON and the matching template
  10. For reflection, guide, insight, and plan, also read docs/reflection-theory.md.
  11. Save daily reports, including daily, to reports/nippo-YYYY-MM-DD.md. Other modes keep reports/{mode}-YYYY-MM-DD.md. Append -Nd when days >…
  12. In daily mode, treat the freshly collected JSON as the only source of truth. Do not read an existing reports/nippo-YYYY-MM-DD.md as input…

What it can do on your machine

Read from SKILL.md and the folder at commit cdd521d. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • cargo
    • gh

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use gh, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Nippo Daily Reports loads about 2.3k tokens when it runs. Until then it costs about 73 tokens; SKILL.md has 1,110 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~73
When it runs · the whole SKILL.md, loaded when a task matches
~2.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from nwiizo/nippo at commit cdd521d, republished under its MIT licence (© nwiizo). 1,110 words, ~2,255 tokens.

Download SKILL.mdSave it as .claude/skills/nippo/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
nippo
description
Generate Japanese daily reports, reflection prompts, guides, reviews, and trend reports from Claude Code, Codex, or opencode work logs. Use when the user asks for nippo, 日報, daily, reflection, guide, report, review, insight, trend, plan, ledger, or wants to summarize recent agent work.

Nippo

Use this skill when the user wants to turn recent Claude Code, Codex, or opencode work into a report under reports/.

Inputs

  • mode: default, daily, brief, reflection, guide, report, review, insight, trend, plan, ledger
  • optional days
  • optional project filter
  • optional source override: claude, codex, opencode, all
  • optional period selector: today, yesterday, this-week, last-week, week-before-last, this-month, last-month, month-before-last

Examples:

  • $nippo daily
  • $nippo daily codex
  • $nippo daily claude
  • $nippo daily all
  • $nippo review 90 codex

Workflow

  1. If the current workspace is this repository, prefer cargo run -q -p nippo -- collect ... so the skill uses the checked-out implementation rather than a potentially stale globally installed nippo.
  2. Otherwise prefer nippo collect ... when nippo is already installed.
  3. If neither is available, stop and tell the user to install nippo.
  4. Treat daily as an alias of the default daily report mode.
  5. Default to --source auto. Override when the user explicitly asks for claude, codex, opencode, or all. The collector removes deterministic prompt noise by default; do not pass --include-prompt-noise during report generation.
  6. For brief, save the summary output directly and stop.
  7. For ledger, do NOT call nippo collect. Run nippo ledger (or cargo run -q -p nippo -- ledger) which scans reports/nippo-*.md (daily reports) for the ## Unclear points section, folds them into reports/ledger.yaml, and prints a CONVERGED / DIVERGENCE-SIGNAL / CONTINUE verdict. Relay that verdict back to the user verbatim; do not over-interpret. If the user asks for agent-config candidates, run nippo ledger --export: it writes recurring General Fix Rules to reports/ledger-export.md. Tell the user these are candidates only — a human must curate them before copying anything into CLAUDE.md / AGENTS.md.
  8. For plan, do NOT call nippo collect. Find the newest reports/nippo-*.md by the date in its filename and read it. Also read reports/ledger.yaml if present and use unresolved recurring General Fix Rules as candidate material. Read docs/templates/plan-template.md and docs/reflection-theory.md, present only 1-3 experiment candidates, leave the choice / reason / first-step fields blank for the user, save to reports/plan-YYYY-MM-DD.md, and report the path.
  9. For other modes, read the collected JSON and the matching template:
  10. For reflection, guide, insight, and plan, also read docs/reflection-theory.md.
  11. Save daily reports, including daily, to reports/nippo-YYYY-MM-DD.md. Other modes keep reports/{mode}-YYYY-MM-DD.md. Append -Nd when days > 1.
  12. In daily mode, treat the freshly collected JSON as the only source of truth. Do not read an existing reports/nippo-YYYY-MM-DD.md as input; overwrite it with the new report.
  13. Ground the daily report header and stats directly in meta and stats, use meta.period for the requested date range, and choose project sections from stats.projects_worked_on in order of message_count.
  14. After generating a daily report (which emits an ## Unclear points section), suggest the user run /nippo ledger to fold today's stuck points into the cumulative streak signal — but do NOT run it automatically.
  15. Finish with the saved path and a concise report summary. Do not suggest collector changes for values already available directly or by simple derivation from the JSON.

Mode Defaults

  • default: --period today
  • daily: alias of default, --period today
  • brief: --period today --format summary
  • reflection: --period today
  • guide: --period today
  • report: --days 7 --stats-only
  • review: --days 90 --stats-only
  • insight: --days 7
  • trend: split the time window into 3 ranges and run 3 summary collections
  • plan: no collection. Reads the newest reports/nippo-*.md and reports/ledger.yaml, writes reports/plan-YYYY-MM-DD.md
  • ledger: no time window. Reads reports/nippo-*.md, writes reports/ledger.yaml, prints a streak verdict. --export writes recurring rules to reports/ledger-export.md
Show full SKILL.md (536 more words)Show less

Rules

  • Data collection must go through nippo collect. Do not reimplement parsing in ad-hoc scripts.
  • Do not use Python for data collection.
  • Use stats as-is. Do not recalculate counters in prose.
  • Write reports in Japanese.
  • Do not recommend books or URLs (hallucination risk). Give concept names and search keywords instead.
  • For reflection and guide, also read the same-day reports/nippo-YYYY-MM-DD.md if it exists.
  • If the first token is neither a mode name, a number, a source selector, nor a period selector, treat it as a project filter and run the default daily mode with --project.
  • Date boundaries follow the machine's local timezone. --days 1 and daily mean the current local calendar day.
  • Treat one token matching claude, codex, opencode, or all as the source selector and pass it through to --source.
  • Treat one token matching today, yesterday, this-week, last-week, week-before-last, this-month, last-month, or month-before-last as a period selector and pass it through to --period. In that case do not add the mode's default --days or default --period today; the explicit --period is the sole time-window flag. For daily, use meta.period.to from the collected JSON as the output filename date instead of computing it from the execution date.
  • Codex report data comes from history.jsonl, state_5.sqlite, and rollout data referenced by rollout_path. Treat logs_2.sqlite as diagnostics only.
  • Codex collection also includes sessions absent from history.jsonl when SQLite references a rollout with user prompts in the requested period. The history file itself is optional; the collector merges duplicate prompts.
  • Codex-derived reports may have sparse assistant/tool metrics. State that explicitly instead of inventing numbers.
  • For daily reports, copy meta.source, meta.total_sessions, stats.projects_worked_on, and stats.tool_frequency from the collected JSON instead of inferring them from an older report.
  • In every mode that prints a date range, use meta.period.from and meta.period.to; do not recalculate the range from the execution date.
  • Cover the activity-heavy projects first. Use stats.projects_worked_on order (= message_count descending) and give the top 3-5 projects their own section before collapsing anything into その他.
  • If you show only a subset of decisions, explicitly state 全N件中M件を記載.
  • The current Claude Code or Codex session is excluded by the collector. Do not manually subtract another session from meta or stats.
  • Sanitize reference links before writing them. Do not copy malformed URL fragments with trailing Japanese text or punctuation.

Improvement Issues

After the report is complete, an Issue may be proposed when report generation repeatedly performs the same deterministic aggregation or formatting, or when one structural problem affects multiple modes. Group symptoms with the same root cause into one coherent Issue instead of proposing one field at a time. Propose at most two Issues per report.

Do not propose an Issue for values already available directly or by simple derivation:

  • Total projects: the length of stats.projects_worked_on
  • Reporting period: meta.period
  • Top tools and main work window: render_helpers
  • Prose or summarization that belongs in a report template

Before proposing, run gh issue list --repo nwiizo/nippo --state all and skip anything already covered or solvable by reading existing fields correctly. If a new Issue is warranted, present a title and body that separate verified facts, affected modes, the shared cause, and the processing that the change would eliminate. Create it with gh issue create --repo nwiizo/nippo only after the user explicitly agrees.

© nwiizo, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in .agents/skills/nippo of nwiizo/nippo.

  • SKILL.md
  • docs

Open the folder on GitHubat commit cdd521d

Compare with similar skills

Nippo Daily Reports next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Nippo Daily Reports compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Nippo Daily Reports this skillnwiizo/nippo180—~2.3kAutomated safety check: PassMIT
Journaldavekilleen/Dex493—~1.8kAutomated safety check: PassMIT
Journalthoreinstein/gemini-obsidian102—~283Automated safety check: PassISC
Claude Code Receipts Reportanthropics/claude-plugins-official38k—~4.3kAutomated safety check: WarnApache-2.0
Daily Journalravila4/claude-adhd-skills157—~2.5kAutomated safety check: PassMIT
DevGlobe Codex TrackerNako0/devglobe-extension216—~554Automated safety check: PassMIT

Similar skills

  • Journal

    davekilleen/Dex

    Toggle journaling or start a morning/evening/weekly journal entry.

    493 GitHub stars~1.8k tokensUpdated today
    Knowledge ManagementAuto-check passed

  • Journal

    thoreinstein/gemini-obsidian

    A skill your agent uses when the user wants to work with their daily note — view today's entry, add a log, do a daily review, or capture quick thoughts.

    102 GitHub stars~283 tokensUpdated 2 mo ago
    Knowledge ManagementAuto-check passed

  • Claude Code Receipts Report

    anthropics/claude-plugins-official

    Official

    Builds a personal Claude Code usage and impact report entirely from local session transcripts and git history, for self-review or justifying usage to a manager.

    38k GitHub stars~4.3k tokensUpdated today
    Productivity & AutomationAuto-check: warnings

  • Daily Journal

    ravila4/claude-adhd-skills

    Draft, organize, or update development journal entries. An agent skill from ravila4/claude-adhd-skills.

    157 GitHub stars~2.5k tokensUpdated 7 mo ago
    Productivity & AutomationAuto-check passed

  • DevGlobe Codex Tracker

    Nako0/devglobe-extension

    DevGlobe activity tracker - appear on the interactive developer globe while coding with Codex. Use $devglobe to setup your API key, set a status message…

    216 GitHub stars~554 tokensUpdated 3 mo ago
    Productivity & AutomationAuto-check passed

  • 21 Day Self Interview

    Forlives/21-day-self-interview

    夜间自我访谈引导（21 天）。扮演资深存在主义心理咨询师，每晚提三个有意义的问题，记录回答并在节点回映，帮助用户看清自己。Use when running a nightly self-inquiry / journaling ritual driven by a scheduled task.

    101 GitHub stars~2k tokensUpdated 4 mo ago
    Productivity & AutomationAuto-check passed

More from nwiizo/nippo

  • Nippo Daily Report Generator

    nwiizo/nippo

    Generates daily reports, reflections, weekly reports, self-reviews and trend insights, in Japanese, from Claude Code, Codex and opencode session logs using the nippo CLI.

    180 GitHub stars~1.7k tokensUpdated 7 days ago
    Auto-check passed

Categories

Questions about Nippo Daily Reports

What does Nippo Daily Reports do?

Turns recent Claude Code, Codex or opencode work into Japanese daily reports, reflections, reviews and trend summaries saved in a reports folder. The skill reads recent agent work and writes a report under `reports/`. A mode argument selects the kind: daily, brief, reflection, guide, report, review, insight, trend, plan or ledger.

When should I use Nippo Daily Reports?

Nippo Daily Reports fits situations like: writing a daily report from what the agent did today; looking back over a week or month of agent work for patterns; summarizing work for one project from a single agent's logs.

How do I install Nippo Daily Reports in Claude Code?

Run `npx skills add nwiizo/nippo --skill nippo -a claude-code`. Or copy the skill folder (.agents/skills/nippo in nwiizo/nippo) into .claude/skills/nippo in your project. Claude Code loads it when a task matches its description.

How do I install Nippo Daily Reports in Codex?

Run `npx skills add nwiizo/nippo --skill nippo -a codex`. Or copy the skill folder (.agents/skills/nippo in nwiizo/nippo) into .agents/skills/nippo in your project. Codex loads it when a task matches its description.

Can I use Nippo Daily Reports in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add nwiizo/nippo --skill nippo -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/nippo, .gemini/skills/nippo, .github/skills/nippo and .opencode/skills/nippo in your project.

What does Nippo Daily Reports need to run?

Going by SKILL.md and its folder, Nippo Daily Reports needs the command-line tools its instructions call (cargo and gh). Our summary lists: The `nippo` CLI, or Cargo inside the nippo repository; Work logs from Claude Code, Codex or opencode on the same machine.

Does Nippo Daily Reports access the network?

SKILL.md contains no URLs. Its commands use gh, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Nippo Daily Reports safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Nippo Daily Reports use?

Nippo Daily Reports is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Nippo Daily Reports use?

About 2.3k tokens (SKILL.md is roughly 9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Nippo Daily Reports?

Skills that share tags, products or a category with Nippo Daily Reports: Journal (davekilleen/Dex, 493 stars), Journal (thoreinstein/gemini-obsidian, 102 stars), Claude Code Receipts Report (anthropics/claude-plugins-official, 38k stars) and Daily Journal (ravila4/claude-adhd-skills, 157 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Nippo Daily Reports?

nwiizo (a GitHub user) maintains it in nwiizo/nippo, which has 180 GitHub stars. The repository holds 2 skills in this directory. The repository was last updated on October 1, 2026.

Source: nwiizo/nippo on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.