debug
Systematic debug workflow: research → orchestrate agents → escalate. Suggest when: stuck after 2-3 attempts, unfamiliar tooling,…
Multi-perspective PR review procedure. Plan a minimal DAG of specialists scoped to what the PR actually touches, synthesize with a critic, post ONE consolidated comment per verbosity tier. Pull before any structured PR review so the methodology stays consistent.
$ npx -y skills add ohdearquant/lionagi --skill pr-review --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/pr-reviewContext preview
The summary Claude sees to decide when to auto-load this skill.
Multi-perspective PR review procedure. Plan a minimal DAG of specialists scoped to what the PR actually touches, synthesize with a critic, post ONE consolidated comment per verbosity tier. Pull before any structured PR review so the methodology stays consistent.
name: pr-review description: > Multi-perspective PR review procedure. Plan a minimal DAG of specialists scoped to what the PR actually touches, synthesize with a critic, post ONE consolidated comment per verbosity tier. Pull before any structured PR review so the methodology stays consistent. allowed-tools: [Bash, Read, Glob, Grep]
A reusable rubric for reviewing a pull request. Works as an orchestrated flow (via the `fanout.submit` or `flow.submit` MCP verbs, or `li o fanout` / `li o flow` from a lionagi checkout) or for a solo reviewer.
Any time you need a structured PR review — OSS contribution evaluation, internal code review, security sign-off, release gate.
| Phase | Action | |-------|--------| | **0. Fetch context** | `gh pr view` + `gh pr diff` → save to `_context/` | | **1. Discovery** | Parallel specialists from closed set (correctness, security, architecture, tests, perf) | | **2. Discussion** | Optional — only if dimensions cross-pollinate | | **3. Critic synthesis** | Reads all findings → `critic_final/final_synthesis.md` | | **4. Post comment** | ONE consolidated comment per verbosity tier |
`APPROVE` | `APPROVE-WITH-FIXES` | `REJECT`
| Verbosity | Content | Safe for | |-----------|---------|----------| | none | don't post, synthesis stays local | any | | brief | ≤10 lines: verdict + top MUST-FIX only | OSS | | substantive | MUST-FIX + SHOULD-FIX with file:line; omit blind-spots | internal repos | | full | complete synthesis including blind spots + coverage | private repos only |
1. Cite line numbers against committed HEAD of the PR branch. 2. Don't invent requirements — only flag if absence is a bug. 3. If runtime context is needed that isn't in the diff, say so. 4. Graceful degradation — note any tool failures in synthesis.
See [specialist-guide.md](specialist-guide.md) for specialist dimensions, severity rubric, synthesis format, MCP and CLI examples, and source code reference.
lionagi is a governed multi-agent orchestration framework. Build agent workflows in Python with typed, inspectable state, or run them straight from the terminal with the li CLI: single agents, parallel fan-outs, and DAG flows where an orchestrator plans
Systematic debug workflow: research → orchestrate agents → escalate. Suggest when: stuck after 2-3 attempts, unfamiliar tooling,…
Fan one task out to multiple independent lionagi workers through `fanout.submit`, optionally synthesizing their answers. Use when…
Plan and run a dependency-aware lionagi DAG through `flow.submit`, including dry runs and reactive follow-up capacity. Use when…
Plan and execute multi-agent workflows through lionagi's orchestration MCP tool: flow.submit (DAG pipelines), fanout.submit…
Run an existing lionagi playbook through the orchestration MCP server and retrieve its background result. Use when a saved…
Author lionagi playbooks — reusable YAML workflow templates that define parametric agent tasks. Playbooks live at…