Denniston — Data & AI agent for Claude Code
PR feedback digest agent.
How to install Denniston
Installs to ~/.claude/agents/jason-witter-legion-denniston.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/jason-witter/Legion/HEAD/agents/denniston.md -o ~/.claude/agents/jason-witter-legion-denniston.md Restart Claude Code, or start a new session, for it to be picked up.
What Denniston does
name: denniston description: PR feedback digest agent. Invoke with structured PR data (reviews, inline comments, CI status, commit history) already fetched by Grace. Triages all reviewer feedback on the user's own pull requests — reads actual source files to validate findings before categorizing — and produces per-PR reports categorizing what needs code changes, what is conversation, and what is blocking. Writes one report per PR to scratch/output/digests/. Does not fetch data, does not modi
Alternatives in Data & AI
- GitHub Issues Specialist — GitHub Issues and Pull Requests specialist optimized for AI-agent workflows 1.5k ★
- AI Maestro: A Development History — From "AI Agents Web Dashboard" to multi-host agent orchestration in 110 days 551 ★
- Reverse DB Architect — Architecte de base de données du reverse DB (Phase 0.B) 193 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Git Steward
Use this agent for anything about the repository's Git/GitHub health and workflow: auditing that no credential
Grace
Source-system fetch specialist. Invoke when the orchestrator needs structured data from a read-only source, su
Orra
Orra — standing AI orchestrator for multi-worktree development. Rides on the Claude Code Agents View. Scans wo
Foley
Backlog triage analyst — stage 1 of the backlog pipeline. Invoke with pre-fetched task data to assess dispatch
Parser
Use when parsing raw story snapshots into structured ParsedStory JSON. Normalizes fetched data into the shared
Ava Issue Triage
Triages open GitHub issues — ranks them by real user impact rather than by label, verifies claims against the
Related Skills
Md Redline
Inline review comments for markdown specs. Built-in MCP server hands feedback directly to your AI agent.
Wishes.Nvim
A Neovim plugin for annotating code with review feedback that any AI coding agent can read and act upon. Inlin
Cdd Process PR
Address the open PR's review feedback: read the review comments for the current branch, triage them, implement