Darwin Skill Optimizer
alchaincyf/darwin-skill
Scores SKILL.md files on a nine-dimension rubric, then improves them in a keep-or-revert loop with independent judge agents, test prompts, git history and human checkpoints.
CEO dashboard with progressive disclosure — 3 tiers: headline (5 lines, default), summary (per-goal detail), deep (full weekly analysis).
$ npx skills add andrew-yangy/gru-ai --skill report -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install andrew-yangy/gru-ai report --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/andrew-yangy/gru-ai.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/report .claude/skills/report && rm -rf skills-srcUse ~/.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/
Install the "report" agent skill from https://github.com/andrew-yangy/gru-ai/tree/main/.claude/skills/report into .claude/skills/report/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "report", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/andrew-yangy/gru-ai/tree/main/.claude/skills/reportType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add andrew-yangy/gru-ai --skill report -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install andrew-yangy/gru-ai report --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/andrew-yangy/gru-ai.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.claude/skills/report .agents/skills/report && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "report" agent skill from https://github.com/andrew-yangy/gru-ai/tree/main/.claude/skills/report into .agents/skills/report/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "report", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add andrew-yangy/gru-ai --skill report -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install andrew-yangy/gru-ai report --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/andrew-yangy/gru-ai.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.claude/skills/report .cursor/skills/report && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "report" agent skill from https://github.com/andrew-yangy/gru-ai/tree/main/.claude/skills/report into .cursor/skills/report/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "report", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/andrew-yangy/gru-ai.git --path .claude/skills/report--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add andrew-yangy/gru-ai --skill report -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install andrew-yangy/gru-ai report --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/andrew-yangy/gru-ai.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.claude/skills/report .gemini/skills/report && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "report" agent skill from https://github.com/andrew-yangy/gru-ai/tree/main/.claude/skills/report into .gemini/skills/report/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "report", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install andrew-yangy/gru-ai reportInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add andrew-yangy/gru-ai --skill report -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/andrew-yangy/gru-ai.git skills-src && mkdir -p .github/skills && cp -r skills-src/.claude/skills/report .github/skills/report && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "report" agent skill from https://github.com/andrew-yangy/gru-ai/tree/main/.claude/skills/report into .github/skills/report/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "report", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add andrew-yangy/gru-ai --skill report -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install andrew-yangy/gru-ai report --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/andrew-yangy/gru-ai.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.claude/skills/report .opencode/skills/report && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "report" agent skill from https://github.com/andrew-yangy/gru-ai/tree/main/.claude/skills/report into .opencode/skills/report/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "report", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
reportCEO dashboard with progressive disclosure — 3 tiers: headline (5 lines, default), summary (per-goal detail), deep (full weekly analysis).
Report is an agent skill from andrew-yangy/gru-ai. CEO dashboard with progressive disclosure — 3 tiers: headline (5 lines, default), summary (per-goal detail), deep (full weekly analysis). Takes an optional argument: headline (default), summary, deep, or legacy aliases daily/weekly.
Its SKILL.md is about 4k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Agent Workflows. It works with Git. The repository describes itself as: Autonomous AI agent team for one-man companies. Context engineering + harness engineering drive a pipeline that brainstorms, builds, reviews, and ships. The licence is MIT.
4 steps, taken from the step headings in SKILL.md.
Read from SKILL.md and the folder at commit 8fba479. It shows what the files ask for, not the result of running them.
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.
Shell commands in SKILL.md call:
gitnpmFrom the folder's file list and the shell code blocks in SKILL.md.
No URLs in SKILL.md. Its commands use git and npm, which can reach the network depending on how they are called.
From URLs in SKILL.md, links to its own repository left out.
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Report loads about 4k tokens when it runs. Until then it costs about 60 tokens; SKILL.md has 732 words of instructions outside code blocks.
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.
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.
The full file from andrew-yangy/gru-ai at commit 8fba479, republished under its MIT licence (© andrew-yangy). 732 words, ~3,967 tokens.
.claude/skills/report/SKILL.md (or your agent's skills folder).Read .claude/agent-registry.json to map roles to agent names. Domain labels in this report use role titles (CTO, CPO, CMO, COO) -- resolve to actual agent names from the registry when generating output.
Generate a CEO report. Tier: $ARGUMENTS (default: headline)
Parse $ARGUMENTS:
What to read depends on the tier. Higher tiers include everything from lower tiers.
.context/directives/*/directive.json — count active goals.context/directives/*/projects/*/project.json — count active/completed/blocked projects, identify recently completed.context/backlog.json — count P0 items needing CEO.context/directives/*.json — count pending directives (filter by status: pending).context/reports/ — scan 3 most recent for unaddressed high-risk follow-ups.context/healthchecks/latest/*.json — quick health status (pass/fail)npm run type-check — build status (pass/fail only, don't show errors here)All headline data, PLUS:
.context/vision.md — for context on what matters.context/preferences.md — CEO standing orders.context/intel/latest/*.json — latest scout outputs per agent~/.conductor/scheduler.json and ~/.conductor/scheduler.log — autopilot status# Recent commits on main (last 24h)
git log --oneline --since="24 hours ago" main
# Active worktrees
git worktree list
# Recently modified context files
find .context/ -name "*.md" -mtime -1 -type f | head -20.context/directives/*/directive.jsonAll summary data, PLUS:
.context/reports/ — compute agent acceptance rates from all reports this week.context/lessons/*.md topic files — any new lessons this week?.context/reports/weekly-*.md)git log --oneline --since="7 days ago" mainThe headline is a 5-line status snapshot. The CEO should know the state of the world in <10 seconds.
# Status — {date}
Shipped: {N} projects completed | Blocked: {N} projects stalled | Needs CEO: {N} decisions pending
Health: {build pass/fail} | {security ok/N issues} | {N active goals}
Top action: {single most important thing the CEO should do next — e.g., "Review 2 high-risk follow-ups from auth-hardening directive" or "Nothing — all clear"}Derivation rules for headline signals:
status changed to "completed" in the last 24h (compare project.json completed timestamps)If the CEO wants more detail: "Run /report summary for per-goal detail or /report deep for full analysis."
# Summary Report — {date}
## External Intelligence (from latest /scout)
- **Technology (CTO)**: {summary + any act_now/this_week items}
- **Product (CPO)**: {summary + competitor moves}
- **Growth (CMO)**: {summary + channel opportunities}
- **Ecosystem (COO)**: {summary + framework updates}
{If no scout data: "No scout data yet — run /scout to gather external intelligence."}
## Internal Health (from latest /healthcheck)
- **Build status**: {type-check pass/fail}
- **Security**: {npm audit summary}
- **Operations**: {stale goals, backlog health}
{If no healthcheck data: "No healthcheck data yet — run /healthcheck."}
## Autopilot
{Read ~/.conductor/scheduler.json and ~/.conductor/scheduler.log}
- **Status**: {enabled/disabled}
- **Budget**: ${spent_today} / ${daily_max} ({pct}%)
- **Recent launches**: {list directives launched today, or "none"}
- **Last check**: {time of most recent check entry}
{If scheduler.json doesn't exist: "Not configured — run scripts/foreman.sh to initialize."}
## Corrections Caught (from recent directives)
{Read recent directive reports for Corrections Caught data.}
- **Last directive**: {name} — {violations_found} violations found, {violations_fixed} fixed
- **Standing Corrections verified**: {count} across {N} initiatives
{If no recent directive data: "No directive executed recently."}
{If violations were found: list each with correction name and resolution}
## What Changed
{Recent commits, summarized by area. Not a raw git log — group and summarize.}
{Active worktrees/branches and their status.}
## Project Inventory
{Read all .context/directives/*/directive.json and .context/directives/*/projects/*/project.json for project inventory.}
### Active Goals ({count})
| Goal | Active Projects | Done | Backlog | Status |
|------|----------------|------|---------|--------|
| {goal title} | {count} ({list names}) | {done_count} | {backlog_count} | {status badge} |
### Partially-Done Alerts
{For each active feature: read project.json embedded tasks, compute completion %. Flag features where:
- completion > 0% but < 100%
- AND last file modification in the feature folder > 7 days ago}
Warning: **{feature name}** ({goal}) — {X}% complete, stale {N} days
Last activity: {date} | Tasks: {completed}/{total}
{If no partially-done alerts: "All active projects are either fresh or complete."}
### Completed But Not Archived
{Projects with status "active" but 100% task completion — should have status updated to "completed"}
- {project name} ({goal}) — 100% complete, still status: active
## What Needs Your Input
{Pending directives in directives/ (filter by status: pending) — list with one-line description each}
{High risk items awaiting decision from recent directives}
{Backlog items flagged for CEO decision}
{If nothing: "Nothing pending — all clear."}
## What's At Risk
{Build errors from type-check, if any}
{Stale active projects with no recent changes}
{Failed/blocked initiatives from recent directives}
{If nothing: "No risks identified."}
## Decision Queue
{Aggregate items needing CEO decision from multiple sources:}
### From Recent Directives
{Scan .context/reports/ for the last 3 directive reports.
For each, check the "Follow-Up Actions" section for high-risk backlogged items.
Cross-reference with .context/backlog.json to see if they've been addressed.}
- **{action}** — Backlogged from {directive name} ({date})
Risk: {high} | Status: {addressed/pending}
### From Backlogs
{Scan all .context/backlog.json for items explicitly marked as needing CEO decision
or items with Priority P0 that are not yet started}
### From Healthchecks
{Read .context/healthchecks/latest/*.json for high-risk findings}
{If nothing: "No pending decisions — all clear."}
## OKR Snapshot
{For each goal with OKRs:}
**{Goal Name}**
- KR-1: {description} — {status} ({metric}: {current} / {target})
- KR-2: ...# Deep Report — Week of {date}
## Executive Summary
{3-5 bullet overview of the week: what shipped, what's in progress, what needs attention}
## External Intelligence (from latest /scout)
- **Technology (CTO)**: {summary + action items}
- **Product (CPO)**: {summary + competitor moves}
- **Growth (CMO)**: {summary + opportunities}
- **Ecosystem (COO)**: {summary + framework updates}
- **Intelligence gathered this week**: {count by urgency level}
{If no scout data: "No scout data yet — run /scout."}
## Internal Health (from latest /healthcheck)
- **Build status**: {type-check pass/fail}
- **Security**: {npm audit summary, CVE count}
- **Operations**: {goal freshness, backlog health}
- **Auto-fixed this period**: {count and summary}
{If no healthcheck data: "No healthcheck data yet — run /healthcheck."}
## Autopilot
{Read ~/.conductor/scheduler.json and ~/.conductor/scheduler.log}
- **Status**: {enabled/disabled}
- **Budget**: ${spent_today} / ${daily_max} ({pct}%)
- **This week**: {count} directives auto-launched, {total estimated cost}
- **Errors this week**: {count, or "none"}
{If scheduler.json doesn't exist: "Not configured — run scripts/foreman.sh to initialize."}
## What Changed This Week
{Commits grouped by area with counts}
{Directives executed and their outcomes}
{Features completed or progressed}
## Project Inventory
{Read all .context/directives/*/directive.json and .context/directives/*/projects/*/project.json for project inventory.}
### Active Goals ({count})
| Goal | Active Projects | Done | Backlog | Status |
|------|----------------|------|---------|--------|
| {goal title} | {count} ({list names}) | {done_count} | {backlog_count} | {status badge} |
### Partially-Done Alerts
{For each active feature: read project.json embedded tasks, compute completion %. Flag features where:
- completion > 0% but < 100%
- AND last file modification in the feature folder > 7 days ago}
Warning: **{feature name}** ({goal}) — {X}% complete, stale {N} days
Last activity: {date} | Tasks: {completed}/{total}
{If no partially-done alerts: "All active projects are either fresh or complete."}
### Completed But Not Archived
{Projects with status "active" but 100% task completion — should have status updated to "completed"}
- {project name} ({goal}) — 100% complete, still status: active
## What Needs Your Input
{Same as summary, but covering the full week}
## What's At Risk
{Same as summary, but covering the full week}
{Trend: are risks increasing or decreasing?}
## Decision Queue
{Aggregate items needing CEO decision from multiple sources:}
### From Recent Directives
{Scan .context/reports/ for the last 3 directive reports.
For each, check the "Follow-Up Actions" section for high-risk backlogged items.
Cross-reference with .context/backlog.json to see if they've been addressed.}
- **{action}** — Backlogged from {directive name} ({date})
Risk: {high} | Status: {addressed/pending}
### From Backlogs
{Scan all .context/backlog.json for items explicitly marked as needing CEO decision
or items with Priority P0 that are not yet started}
### From Healthchecks
{Read .context/healthchecks/latest/*.json for high-risk findings}
{If nothing: "No pending decisions — all clear."}
## What Shifted This Week
{Compare current project inventory against what was reported in the most recent saved weekly report
in .context/reports/weekly-*.md}
### New Work Started
- {feature name} ({goal}) — started this week
### Completed This Week
- {project name} ({goal}) — status changed to completed
### Went Stale (no activity >7 days)
- {feature name} ({goal}) — last activity {date}
### Priority Changes
{Any backlog items that were promoted from /scout this week}
{If no previous weekly report exists: "First weekly report — no comparison available."}
## OKR Progress
{Full OKR breakdown per goal — same as summary but with week-over-week changes if prior report exists}
## Team Performance
{From recent scout reports in .context/intel/latest/:}
- **Proposals this week**: {count} ({count} approved, {count} rejected, {count} deferred)
- **By agent**:
- CTO: {proposed} proposed, {accepted} accepted ({rate}%)
- CPO: {proposed} proposed, {accepted} accepted ({rate}%)
- CMO: {proposed} proposed, {accepted} accepted ({rate}%)
- COO: {proposed} proposed, {accepted} accepted ({rate}%)
{From directive reports:}
- **Directives completed**: {count}
- **Initiatives**: {completed}/{total} ({rate}%)
- **Build success rate**: {pass}/{total} type-checks passed
{From directive reports this week:}
- **Corrections enforced**: {total corrections checked across all directives}
- **Violations caught**: {count} (by role: CTO {N}, CPO {N}, COO {N})
- **Violation types**: {which standing corrections were violated most}
## Corrections Caught This Week
{Aggregate corrections_check data from all directive reports this week:}
| Correction | Directive | Caught By (role) | Resolution |
|------------|-----------|-------------------|------------|
| {correction} | {directive} | {reviewer role} | {fixed/noted} |
- **Trend**: {more/fewer/same violations as last week}
{If no violations: "Clean week — all standing corrections respected across all directives."}
## Lessons Learned
{New entries in .context/lessons/ topic files from this week}
{If none: "No new lessons captured this week."}
## Recommendations
{COO-style operational recommendations:}
- What should the CEO focus on next week?
- Any goals that need re-prioritization?
- Any process improvements to consider?Output the report directly to the CEO. Do NOT write it to a file unless the CEO asks.
For deep tier reports, also offer: "Would you like me to save this report to .context/reports/weekly-{date}.md?"
After any tier, remind: "Drill down with /report summary or /report deep for more detail." (Skip this line if already at deep tier.)
| Situation | Action |
|---|---|
| No OKR files exist | Skip OKR section, note "No OKRs tracked yet" |
| No scout data (intel/latest/) | Skip external intelligence section, note "No scout data yet — run /scout" |
| No healthcheck data (healthchecks/latest/) | Skip internal health section, note "No healthcheck data yet — run /healthcheck" |
| No scout data exists | Skip team performance section, note "No scout/patrol data yet" |
| Type-check fails to run | Note the error, skip build health section |
| No recent commits | Note "No commits in {timeframe}" |
| No directive reports | Skip directive outcomes section |
| No active projects found | Note "No active projects tracked yet" in Project Inventory |
| No tasks in a project.json | Skip that feature in completion calculations |
| No previous weekly report for comparison | Note "First weekly report — no comparison available" in Shift Tracking |
| No conductor reports for Decision Queue | Note "No directive history to scan" |
| No healthcheck data for Decision Queue | Skip "From Healthchecks" subsection |
| No ~/.conductor/scheduler.json | Note "Autopilot not configured" in Autopilot section |
| No ~/.conductor/scheduler.log | Note "No scheduler activity yet" |
npm run build, no npm run lint)© andrew-yangy, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Just SKILL.md in .claude/skills/report of andrew-yangy/gru-ai.
Open the folder on GitHubat commit 8fba479
Report 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.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Report this skillandrew-yangy/gru-ai | 155 | — | ~4k | Automated safety check: Pass | MIT | |
| Darwin Skill Optimizeralchaincyf/darwin-skill | 6.2k | 1 repos | ~4.7k | Automated safety check: Pass | MIT | |
| CodeGraph Agent Evalcolbymchenry/codegraph | 74k | — | ~950 | Automated safety check: Pass | MIT | |
| Neat-Freak Knowledge CloseoutKKKKhazix/khazix-skills | 21k | — | ~1.9k | Automated safety check: Pass | MIT | |
| O2 Review Loopopenobserve/openobserve | 22k | — | ~3.7k | Automated safety check: Pass | AGPL-3.0 | |
| Beads Task Memorygastownhall/beads | 28k | — | ~1.2k | Automated safety check: Pass | MIT |
alchaincyf/darwin-skill
Scores SKILL.md files on a nine-dimension rubric, then improves them in a keep-or-revert loop with independent judge agents, test prompts, git history and human checkpoints.
colbymchenry/codegraph
Benchmarks how much CodeGraph helps a coding agent on a real repository, comparing runs with and without it for a chosen local or published version.
KKKKhazix/khazix-skills
Brings project docs, agent rule files, authorized memory and leftover workspace files back in line with what the code and runtime actually do at the end of a work session.
openobserve/openobserve
Splits a change into planner, coder and independent reviewer roles: you confirm a spec, a subagent implements it, and a separate reviewer checks each round's local WIP commit.
gastownhall/beads
Tracks multi-session work with dependencies in the bd issue tracker so the agent can find ready tasks and recover its context after conversation compaction.
slopus/happy
Searches past Claude Code, Codex and Cursor sessions and summarizes what was worked on, tried or decided, using extraction scripts instead of reading raw logs.
andrew-yangy/gru-ai
Provides comprehensive code review guidance for React 19, Vue 3, Rust, TypeScript, Java, Python, and C/C++.
andrew-yangy/gru-ai
Full website SEO audit with parallel subagent delegation. An agent skill from andrew-yangy/gru-ai.
andrew-yangy/gru-ai
Structured brainstorm — from quick Socratic refinement to full C-suite strategy sessions.
andrew-yangy/gru-ai
Internal codebase and operations health check — the CTO scans technical health, the COO checks operational health.
andrew-yangy/gru-ai
Pipeline end-to-end smoke test -- creates a trivial directive, runs it through /directive, validates every pipeline step, and reports pass/fail with evidence.
andrew-yangy/gru-ai
Execute work through the directive pipeline — evaluate, plan, cast agents, build, review, and report.
Works with
Categories
CEO dashboard with progressive disclosure — 3 tiers: headline (5 lines, default), summary (per-goal detail), deep (full weekly analysis). Report is an agent skill from andrew-yangy/gru-ai. CEO dashboard with progressive disclosure — 3 tiers: headline (5 lines, default), summary (per-goal detail), deep (full weekly analysis).
Report fits situations like: agent Workflows work in your project.
Run `npx skills add andrew-yangy/gru-ai --skill report -a claude-code`. Or copy the skill folder (.claude/skills/report in andrew-yangy/gru-ai) into .claude/skills/report in your project. Claude Code loads it when a task matches its description.
Run `npx skills add andrew-yangy/gru-ai --skill report -a codex`. Or copy the skill folder (.claude/skills/report in andrew-yangy/gru-ai) into .agents/skills/report in your project. Codex loads it when a task matches its description.
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add andrew-yangy/gru-ai --skill report -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/report, .gemini/skills/report, .github/skills/report and .opencode/skills/report in your project.
Going by SKILL.md and its folder, Report needs the command-line tools its instructions call (git and npm).
SKILL.md contains no URLs. Its commands use git and npm, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
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.
Report is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
About 4k tokens (SKILL.md is roughly 16k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
Skills that share tags, products or a category with Report: Darwin Skill Optimizer (alchaincyf/darwin-skill, 6.2k stars), CodeGraph Agent Eval (colbymchenry/codegraph, 74k stars), Neat-Freak Knowledge Closeout (KKKKhazix/khazix-skills, 21k stars) and O2 Review Loop (openobserve/openobserve, 22k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
andrew-yangy (a GitHub user) maintains it in andrew-yangy/gru-ai, which has 155 GitHub stars. The repository holds 9 skills in this directory. The repository was last updated on March 11, 2026.
Source: andrew-yangy/gru-ai on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.