Load PR Comments
NeoLabHQ/context-engineering-kit
A skill your agent uses to load open/unresolved PR review comments then aggregate them as tasks in .specs/comments/.md for parallel agents to fix.
A skill your agent uses when a pr-council-review (PR or branch mode) result needs independent corroboration from a different model family before trusting a GREEN verdict — high-stakes…
Install LLM Council
$ npx skills add ffroliva/gflow-cli --skill llm-council -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install ffroliva/gflow-cli llm-council --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/ffroliva/gflow-cli.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/llm-council .claude/skills/llm-council && 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 "llm-council" agent skill from https://github.com/ffroliva/gflow-cli/tree/develop/skills/llm-council into .claude/skills/llm-council/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "llm-council", 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/ffroliva/gflow-cli/tree/develop/skills/llm-councilType 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 ffroliva/gflow-cli --skill llm-council -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install ffroliva/gflow-cli llm-council --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/ffroliva/gflow-cli.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/llm-council .agents/skills/llm-council && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "llm-council" agent skill from https://github.com/ffroliva/gflow-cli/tree/develop/skills/llm-council into .agents/skills/llm-council/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "llm-council", 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 ffroliva/gflow-cli --skill llm-council -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install ffroliva/gflow-cli llm-council --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/ffroliva/gflow-cli.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/llm-council .cursor/skills/llm-council && 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 "llm-council" agent skill from https://github.com/ffroliva/gflow-cli/tree/develop/skills/llm-council into .cursor/skills/llm-council/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "llm-council", 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/ffroliva/gflow-cli.git --path skills/llm-council--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 ffroliva/gflow-cli --skill llm-council -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install ffroliva/gflow-cli llm-council --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/ffroliva/gflow-cli.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/llm-council .gemini/skills/llm-council && 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 "llm-council" agent skill from https://github.com/ffroliva/gflow-cli/tree/develop/skills/llm-council into .gemini/skills/llm-council/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "llm-council", 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 ffroliva/gflow-cli llm-councilInstalls 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 ffroliva/gflow-cli --skill llm-council -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/ffroliva/gflow-cli.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/llm-council .github/skills/llm-council && 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 "llm-council" agent skill from https://github.com/ffroliva/gflow-cli/tree/develop/skills/llm-council into .github/skills/llm-council/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "llm-council", 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 ffroliva/gflow-cli --skill llm-council -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install ffroliva/gflow-cli llm-council --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/ffroliva/gflow-cli.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/llm-council .opencode/skills/llm-council && 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 "llm-council" agent skill from https://github.com/ffroliva/gflow-cli/tree/develop/skills/llm-council into .opencode/skills/llm-council/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "llm-council", 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.
Facts
- Skill name
llm-council- GitHub stars
- 269
- Token cost
- ~2k tokens
- SKILL.md length
- 1,018 words
- Files
- 1
- Skills in repo
- 17
- Repo updated
- First seen
- Licence
- MIT
At a glance
A skill your agent uses when a pr-council-review (PR or branch mode) result needs independent corroboration from a different model family before trusting a GREEN verdict — high-stakes…
- Works in 7 steps: Resolve tier → tool list. → Probe every resolved tool in parallel,… → Dispatch pr-council-review (unchanged)… → …
- A pr-council-review (PR
- SKILL.md covers Overview, When to Use, Quick Reference — Tiers and Tool Registry, plus 3 more sections
- Calls codex and git
What it does
LLM Council is an agent skill from ffroliva/gflow-cli. Use when a pr-council-review (PR or branch mode) result needs independent corroboration from a different model family before trusting a GREEN verdict — high-stakes, security-sensitive, or architecturally significant reviews where same-model-family Claude subagents might share a blind spot. Also use when the user asks for "external", "second opinion", "cross-model", or names codex/antigravity (agy) or another external CLI coding agent alongside a review.
Its SKILL.md is about 2k 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, covering Pull requests, Subagents and AI video generation. The repository describes itself as: Drive Google Flow from the command line: Veo video and Imagen images, scripted, batched and pipeline-ready. Ships an MCP server so coding agents can drive it too, giving you and… The licence is MIT.
When your agent uses it
- A pr-council-review (PR
- Branch mode) result needs independent corroboration from a different model family before trusting a GREEN verdict — high-stakes
- Security-sensitive
- Architecturally significant reviews where same-model-family Claude subagents might share a blind spot
Example prompts
- “external”
- “second opinion”
- “cross-model”
- “/llm-council”
Workflow steps
7 steps, taken from the first numbered list in SKILL.md.
- Resolve tier → tool list.
- Probe every resolved tool in parallel, short timeout (~10-15s). A tool that doesn't respond is excluded from this round — name it in the…
- Dispatch pr-council-review (unchanged) for the internal dimension council.
- In parallel, dispatch each surviving external tool via its registry recipe, backgrounded.
- Fold each returning verdict into the same synthesis table pr-council-review produces — same GREEN/YELLOW/RED vocabulary, tagged by source…
- Any finding — internal or external — that warrants a fix gets applied, then re-verified against the specific dimension/tool that flagged…
- Report: pr-council-review's existing shape, plus an "External tools" line noting which ran / were skipped / failed and why.
What it can do on your machine
Read from SKILL.md and the folder at commit d44abc8. 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:
codexgit
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use git, 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
LLM Council loads about 2k tokens when it runs. Until then it costs about 118 tokens; SKILL.md has 1,018 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.
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 ffroliva/gflow-cli at commit d44abc8, republished under its MIT licence (© ffroliva). 1,018 words, ~1,959 tokens.
Download SKILL.mdSave it as
.claude/skills/llm-council/SKILL.md (or your agent's skills folder).- name
- llm-council
- description
- Use when a pr-council-review (PR or branch mode) result needs independent corroboration from a different model family before trusting a GREEN verdict — high-stakes, security-sensitive, or architecturally significant reviews where same-model-family Claude subagents might share a blind spot. Also use when the user asks for "external", "second opinion", "cross-model", or names codex/antigravity (`agy`) or another external CLI coding agent alongside a review.
llm-council — external-tools review layer
Overview
Wraps pr-council-review (unchanged) and adds a layer of external CLI coding agents (codex, plus Antigravity — the agy harness) as additional independent reviewers, then folds their verdicts into the same synthesis. Internal Claude subagents are independent per-dimension but share one model family's blind spots — a phrasing choice, a Windows-vs-POSIX nuance, or a syntax error that reads fine to one Claude reviewer reads fine to all of them. A different model family catches a different error distribution. Confirmed live: on one review, external tools caught 6 real, distinct issues (a wording-accuracy bug, a wrong test count, a Windows-only test-triviality nuance, a test-isolation gap, a missing test, a malformed markdown fence) that 12 internal Claude-subagent dispatches (6 dimensions × 2 rounds) had all missed.
When to Use
- Any
pr-council-reviewrun (PR# mode or branch mode) where the artifact is high-stakes enough to want a second, differently-biased opinion before calling it GREEN. - Not needed for a quick spot-check or draft iteration — use
/review(single-agent) for that;pr-council-reviewalone for a normal-stakes PR.
Quick Reference — Tiers
| Tier | Internal (pr-council-review) | External tools |
|---|---|---|
small (default) | ✅ full dimension council | none — identical to running pr-council-review directly |
medium | ✅ | codex |
high | ✅ | codex + Antigravity (agy) |
Tier controls tool breadth, not review rounds. Fix → re-verify → repeat until GREEN (or a round cap) happens at every tier — that's how council review works, not a tier knob.
Tool Registry
Fixed, tested invocation recipes. Do not improvise a command for a listed tool — the "obvious" invocation is often a trap (see codex below).
codex
- NEVER
codex review. Its built-in prompt has gotten stuck in a self-inflicted loop reading skill files via a malformed PowerShell command, then retrying the identical broken command for 20+ minutes with zero progress. Confirmed reproducible on a clean retry. - Use:
codex exec -s read-only -C <absolute-repo-dir> --skip-git-repo-check "<direct, fully self-contained prompt>". - Probe:
codex --version(near-instant; confirms binary health only, not auth/quota). - Timeout budget: real calls run 10-20 min at default (
xhigh) reasoning effort. Always background it — never block synchronously. - Orphan risk: a killed/timed-out
codex execcan leavecodex.exe/codex-code-mode-host.exe/ sandbox-helper processes running on Windows. After any kill, verify viatasklist/psthat the named PIDs are actually gone before retrying — a retry racing an orphan still writing the same output path silently corrupts the result.
Antigravity (agy)
- What it is: Google's Antigravity harness, invoked via
agy. It supplies thehigh-tier's second, different-model-family opinion. - Working recipe (verified 2026-08-27):
agy --model <model> --mode plan --dangerously-skip-permissions --add-dir <absolute-repo-dir> --print-timeout 15m -p "<prompt>".--mode planis what makes the permission flag acceptable: the agent gets reads, not writes. Verify the worktree is unmodified afterwards (git status) — on the run that produced this note, it was. - Do NOT use
--agent <gsd-*> --new-project. That was the previous recipe and it terminates with a bareError: Agent execution terminated due to error.— this was misdiagnosed here as quota exhaustion for months. - Probe:
agy models— notagy --version. The version check only proves the binary exists;agy modelsexercises auth and returns the model list (Gemini 3.x Pro/Flash, Claude, GPT-OSS), which is what you actually need to know. - The failure mode is permissions, not quota. Headless mode cannot prompt, so tool requests are auto-denied and the run returns no output at all:
a tool required the "command" permission that headless mode cannot prompt for, so it was auto-denied. Three separate runs failed three different ways before this surfaced; each error named the next problem, so read the actual message rather than assuming quota. - If it's genuinely unavailable: don't silently retry past its probe. Suggest installing it, or substitute another external CLI coding agent, and continue best-effort with whatever did return; never block the whole round on it. Note that the
geminiCLI is not a valid substitute for individual accounts — it now returnsIneligibleTierError: This client is no longer supported for Gemini Code Assist for individuals, and it fails that way after passing a--versionprobe.
Show full SKILL.md (369 more words)Show less
Dispatch Flow
- Resolve tier → tool list.
- Probe every resolved tool in parallel, short timeout (~10-15s). A tool that doesn't respond is excluded from this round — name it in the report, don't just drop it silently. The probe only catches binary-health failures (not installed, hung shell); a quota-exhausted tool can still pass the probe and fail on the real call — that's what the dispatch-layer disclosure step below is for.
- Dispatch
pr-council-review(unchanged) for the internal dimension council. - In parallel, dispatch each surviving external tool via its registry recipe, backgrounded.
- Fold each returning verdict into the same synthesis table
pr-council-reviewproduces — same GREEN/YELLOW/RED vocabulary, tagged by source (e.g.D3 (internal)vscodex (external)). A tool that fails or times out after passing its probe is dropped from this round with an explicit note in the report. Never silently drop, never block the whole round on one flaky tool. - Any finding — internal or external — that warrants a fix gets applied, then re-verified against the specific dimension/tool that flagged it (not necessarily the whole pool again).
- Report:
pr-council-review's existing shape, plus an "External tools" line noting which ran / were skipped / failed and why.
Common Mistakes
| Mistake | Fix |
|---|---|
Running codex review because it sounds like the obvious subcommand for a review task | Use codex exec -s read-only -C <dir> --skip-git-repo-check "<prompt>" — see registry |
| Dispatching the real (slow) external call before probing | Probe first, short timeout — a dead tool costs 10-20 min discovered late vs. ~15s discovered early |
| Treating one failed external tool as a reason to abandon the whole external layer | Best-effort: drop that tool for this round, disclose it, keep going with whatever did return |
| Blocking synchronously on an external tool call | Always background it — internal dimensions and other external tools shouldn't wait |
| Retrying a timed-out tool without checking for orphaned processes first | tasklist/ps check + explicit kill before any retry against the same output path |
Silently downgrading to codex-only when agy is unavailable | Disclose the drop and suggest installing Antigravity or substituting another external CLI agent — don't hide the reduced coverage |
Cross-References
REQUIRED SUB-SKILL: the internal council dispatch is pr-council-review (skills/pr-council-review/SKILL.md) — this skill does not reimplement dimension detection, synthesis rules, or report shape, it wraps them.
© ffroliva, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in skills/llm-council of ffroliva/gflow-cli.
Open the folder on GitHubat commit d44abc8
Compare with similar skills
LLM Council 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 |
|---|---|---|---|---|---|---|
| LLM Council this skillffroliva/gflow-cli | 269 | — | ~2k | Automated safety check: Pass | MIT | |
| Load PR CommentsNeoLabHQ/context-engineering-kit | 1.8k | — | ~2.1k | Automated safety check: Pass | GPL-3.0 | |
| Multi Agent PR Reviewbbartling/open-fdd | 173 | — | ~1.4k | Automated safety check: Pass | Custom licence | |
| Gh Issuestrpc-group/trpc-agent-go | 1.9k | 8 repos | ~8.7k | Automated safety check: Pass | Apache-2.0 | |
| Aidd PRparalleldrive/aidd | 384 | — | ~1k | Automated safety check: Pass | MIT | |
| GitHub Review Iterationprisma/orm | 48k | — | ~2.2k | Automated safety check: Pass | Apache-2.0 |
Similar skills
Multi Agent PR Review
bbartling/open-fdd
A skill your agent uses for rigorous pull request, branch, patch, or diff review.
Gh Issues
trpc-group/trpc-agent-go
Fetch GitHub issues, spawn sub-agents to implement fixes and open PRs, then monitor and address PR review comments.
Aidd PR
paralleldrive/aidd
Triage PR review comments, resolve already-addressed threads, and delegate /aidd-fix prompts for remaining issues.
- Official
GitHub Review Iteration
prisma/orm
Runs a loop on a GitHub pull request: fetch review state, triage comments into actions, implement them and resolve threads, repeating until nothing actionable is left.
Cherry Studio PR Review
CherryHQ/cherry-studio
Reviews Cherry Studio branches, pull requests, commits, files and docs against the project's own architecture, naming, API-boundary and UI rules, report-only by default.
More from ffroliva/gflow-cli
All 17 skills in this repoGflow CLI
ffroliva/gflow-cli
A skill your agent uses when the user wants to drive Google Flow (Veo image-to-video, Veo text-to-video, Imagen / Nano Banana image generation) from the terminal or a script — including…
Issue Assessment
ffroliva/gflow-cli
A skill your agent uses when triaging a GitHub issue for gflow-cli — a reporter's bug claim, a freshly-filed issue, or deciding whether and how to act on one.
Issue Resolve
ffroliva/gflow-cli
A skill your agent uses when an assessed gflow-cli issue (verdict CONFIRMED-BUG or LIKELY-BUG) has localized, verifiable scope and should be driven to a fix.
Live Verify
ffroliva/gflow-cli
Two-part gate for gflow-cli feature/fix work. An agent skill from ffroliva/gflow-cli.
Video Production
ffroliva/gflow-cli
A skill your agent uses when the user wants a finished video out of gflow rather than a single clip — a scripted scene, a talking-head or dialogue piece, an explainer, a product montage, a story…
Check
ffroliva/gflow-cli
Auto-fix lint and formatting, then report types and tests. An agent skill from ffroliva/gflow-cli.
Related
Questions about LLM Council
What does LLM Council do?
A skill your agent uses when a pr-council-review (PR or branch mode) result needs independent corroboration from a different model family before trusting a GREEN verdict — high-stakes…. LLM Council is an agent skill from ffroliva/gflow-cli. Use when a pr-council-review (PR or branch mode) result needs independent corroboration from a different model family before trusting a GREEN verdict — high-stakes, security-sensitive, or architecturally significant reviews where same-model-family Claude subagents might share a blind spot.
When should I use LLM Council?
LLM Council fits situations like: A pr-council-review (PR; branch mode) result needs independent corroboration from a different model family before trusting a GREEN verdict — high-stakes; security-sensitive; architecturally significant reviews where same-model-family Claude subagents might share a blind spot.
How do I install LLM Council in Claude Code?
Run `npx skills add ffroliva/gflow-cli --skill llm-council -a claude-code`. Or copy the skill folder (skills/llm-council in ffroliva/gflow-cli) into .claude/skills/llm-council in your project. Claude Code loads it when a task matches its description.
How do I install LLM Council in Codex?
Run `npx skills add ffroliva/gflow-cli --skill llm-council -a codex`. Or copy the skill folder (skills/llm-council in ffroliva/gflow-cli) into .agents/skills/llm-council in your project. Codex loads it when a task matches its description.
Can I use LLM Council 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 ffroliva/gflow-cli --skill llm-council -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/llm-council, .gemini/skills/llm-council, .github/skills/llm-council and .opencode/skills/llm-council in your project.
What does LLM Council need to run?
Going by SKILL.md and its folder, LLM Council needs the command-line tools its instructions call (codex and git).
Does LLM Council access the network?
SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Is LLM Council 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 LLM Council use?
LLM Council 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 LLM Council use?
About 2k tokens (SKILL.md is roughly 7.8k 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 LLM Council?
Skills that share tags, products or a category with LLM Council: Load PR Comments (NeoLabHQ/context-engineering-kit, 1.8k stars), Multi Agent PR Review (bbartling/open-fdd, 173 stars), Gh Issues (trpc-group/trpc-agent-go, 1.9k stars) and Aidd PR (paralleldrive/aidd, 384 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains LLM Council?
ffroliva (a GitHub user) maintains it in ffroliva/gflow-cli, which has 269 GitHub stars. The repository holds 17 skills in this directory. The repository was last updated on October 8, 2026.
Source: ffroliva/gflow-cli on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.