remy-insight
Deep repository analysis with multi-agent parallel perspectives.
Skill by MyPeacefulValentine
Chalk Labs
Need help deploying remy-insight?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/mypeacefulvalentine-remy-insight
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: research, agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Python package
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/remy-insight
Tools it pre-approves 8
From allowed-tools in SKILL.md: the agent may use these without asking you each time.
Read,Grep,Glob,Bash,PowerShell,Write,AskUserQuestion,Agent
What's inside
- Skill name
remy-insight- Description
- Deep repository analysis with multi-agent parallel perspectives. Requires /init + /remy-index as prerequisites. Produces structured research reports.
- License
- MIT
- Allowed tools
- Read,, Grep,, Glob,, Bash,, PowerShell,, Write,, AskUserQuestion,, Agent
- Instructions
- 537 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
- Pre-approves some shell commands
Agent Skills spec issues
- body: 537 lines (spec recommends under 500)
Files (18)
- README.md
- README_zh.md
- SKILL.md
- output_schema.json
- angle_architecture.md · prompts/
- angle_custom.md · prompts/
- angle_improvement.md · prompts/
- angle_innovation.md · prompts/
- angle_robustness.md · prompts/
- render.py
- agent_finding.json · schemas/
- _base.md.j2 · templates/
- section_architecture.md.j2 · templates/
- section_custom.md.j2 · templates/
- section_executive.md.j2 · templates/
- section_improvement.md.j2 · templates/
- section_innovation.md.j2 · templates/
- section_robustness.md.j2 · templates/
SKILL.md (first lines)
# Repository Insight Protocol
Multi-dimensional, multi-agent deep semantic analysis skill. Consumes logic_index.db (SQLite) to analyze repository architecture, identify improvement opportunities, and verify documentation-code consistency.
**Relationship to other skills:**
| Skill | Relationship |
| :--- | :--- |
| remy-index | Data source. Insight consumes logic_index.db |
| remy-reposcout | Upstream. reposcout does shallow recon → user decides to clone → insight does deep research |
| remy-audit | No overlap. audit targets incremental diffs; insight targets whole-repo / focused research |
| remy-secure | Partial overlap (robustness dimension). insight gives high-level observation; secure does rule-driven scanning |
## 0. Configuration
| Environment Variable | Default | Description |
| :--- | :--- | :--- |
| `INSIGHT_DEFAULT_DEPTH` | `standard` | Depth level when `--depth` is not specified |
| `INSIGHT_MAX_CUSTOM_ANGLES` | `2` | Maximum user-defined custom analysis angles |
| `INSIGHT_MAX_AGENTS` | `40` | Hard cap on total agents per run (analysis + adversarial combined) |
## External Files
> **Path Convention**: All paths below are relative to `~/.claude/`. Use `Read("~/.claude/skills/remy-insight/...")` to access them.
| File | Purpose |
| :--- | :--- |
| `skills/remy-insight/output_schema.json` | Report output JSON Schema |
| `skills/remy-insight/schemas/agent_finding.json` | Single agent output Schema |
| `skills/remy-insight/prompts/angle_architecture.md` | Architecture quality assessment prompt |
| `skills/remy-insight/prompts/angle_improvement.md` | Improvement opportunity identification prompt |
| `skills/remy-insight/prompts/angle_robustness.md` | Security & robustness assessment prompt |
| `skills/remy-insight/prompts/angle_innovation.md` | Technical innovation identification prompt |
| `skills/remy-insight/prompts/angle_custom.md` | User-defined custom angle framework template |
| `skills/remy-insight/templates/_base.md.j2` | Report shell (metadata, section loop, methodology) |
| `skills/remy-insight/templates/section_executive.md.j2` | Executive Summary section |
| `skills/remy-insight/templates/section_architecture.md.j2` | Architecture analysis section |
| `skills/remy-insight/templates/section_innovation.md.j2` | Innovation analysis section |
| `skills/remy-insight/templates/section_improvement.md.j2` | Improvement roadmap section |
| `skills/remy-insight/templates/section_robustness.md.j2` | Robustness analysis section |
| `skills/remy-insight/templates/section_custom.md.j2` | Custom dimension section |
| `skills/remy-insight/render.py` | Template rendering + section assembly + cross-reference annotation + CLI entry point |
## Optional Dependency: Jinja2
`render.py` attempts `import jinja2`. If unavailable, all templates are rendered via built-in string formatting. Jinja2 is optional; install it with `pip install jinja2`.
---
## Phase 0: Input Parsing & Intent Extraction
```
/remy-insight [mode] [args...] [--depth light|standard|deep]
```
### Parsing Rules
1. First argument matches `global` → **global** mode
2. First argument matches `focus` → **focus** mode, second argument is `<topic>`. Optional `--with <section_name>` to append additional sections.
3. First argument is a file path ending in `.md` or `.tex` → **compare** mode
4. First argument matches `compare` → **compare** mode, second argument is `<doc_path>`
5. All other non-empty input → use `AskUserQuestion` to present available modes and let user choose. Do NOT guess intent.
6. No argument → use `AskUserQuestion` to present available modes with descriptions.
### Mode-Section Matrix
| Mode | Default Sections | Description |
| :--- | :--- | :--- |
| `global` | executive, architecture, innovation, improvement, robustness | Full repository analysis |
| `focus <topic>` | executive, architecture, improvement | Focused module/subsystem analysis |
| `compare <doc_path>` | executive, doc_consistency, architecture | Document vs code consistency check |
For `focus` mode, `--with <section>` appends additional sections (e.g., `focus auth --with robustness`).
### Depth Levels
| Depth | Analysis Agents | Multi-Instance | Adversarial Verification |
| :--- | :--- | :--- | :--- |
| `light` | 2 (architecture + improvement only) | No | No |
| `standard` | Per active section count (2-5) | No | Yes — single-refute per issue-severity finding |
| `deep` | Per active section × 2-3 instances | Yes (2-3 per angle) | Yes — 3-vote per concern/issue finding (≥2 consensus adopts) |
**Default depth**: Read `INSIGHT_DEFAULT_DEPTH` from environment. Falls back to `standard`.
---
## Phase 1: Prerequisite Validation
Execute all checks sequentially. HALT on any failure.
### 1.1 CLAUDE.md Check
Run `Bash("test -f CLAUDE.md && echo EXISTS || echo MISSING")` (or PowerShell equivalent on Windows).
- **MISSING**: Print "CLAUDE.md not found. Run `/init` first." → HALT.
### 1.2 Logic Index Check
Run `Bash("test -f .claude/logic_index.db && echo EXISTS || echo MISSING")`.
- **MISSING**: Use `AskUserQuestion`:
> "`.claude/logic_index.db` does not exist. Run `/remy-index` to generate? This is required for remy-insight."
- **Yes**: Invoke `remy-index` skill, then continue.
- **No**: HALT.
### 1.3 Freshness Check (Three-Layer)
**Layer 1 — Dirty file tracker**:
Check if `.claude/logic_index_dirty` exists and is non-empty.
- Non-empty → Flag as `DIRTY`.
**Layer 2 — File set diff**:
`Glob` the current source files matching parser extensions (`.py`, `.c`, `.cpp`, `.h`, `.ts`, `.tsx`). Compare against file paths from `SELECT path FROM files` in `.claude/logic_index.db`. Detect additions and deletions.
- Any diff → Flag as `STALE_FILES`.
**Layer 3 — Modification time sampling**:
Query `SELECT value FROM meta WHERE key='last_updated'` from `logic_index.db`. For up to 10 randomly selected indexed files, compare file modification time (mtime) against `last_updated`. Use seconds-level precision for cross-platform compatibility.
- Any file mtime > last_updated → Flag as `STALE_MTIME`.From MyPeacefulValentine/Remy-CC, MIT.
Signals
- Stars
- 28
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 17, 2026 (1 month ago)
- Last release
- Oct 18, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/MyPeacefulValentine/Remy-CC
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get remy-insight running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
remy-insight skill. Agent Mart. https://agentmart.dev/skills/mypeacefulvalentine-remy-insight (last updated 2026-10-24)