graph-auditor
Chain-of-reasoning (spoken, ends in a decision) — an orphan node has no reason to exist, an ambiguous arrow is a lie, and invisible feedback is a hidden loop…
Skill by sancovp
Chalk Labs
Need help deploying graph-auditor?
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/sancovp-graph-auditor
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
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. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/graph-auditor
What's inside
- Skill name
graph-auditor- Description
- Chain-of-reasoning (spoken, ends in a decision) — an orphan node has no reason to exist, an ambiguous arrow is a lie, and invisible feedback is a hidden loop — find them and repair the graph.
- License
- MIT
- Instructions
- 23 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
This is a **chain of reasoning (CoR)** — a SPOKEN, paragraphical reasoning chain that ends in a DECISION. Say it as a paragraph; keep the moves in order. ## CoR (custom syntax) ```rune [Graph] ⇒ [OrphanNodes] ⇒ [AmbiguousEdges] ⇒ [AbstractionCollisions] ⇒ [MissingTransitions] ⇒ [FeedbackVisibility] ⇒ |RepairedGraph| ``` ## Say it as a paragraph, in order State your reasoning as ONE paragraph that performs, in order, 7 moves: Graph, OrphanNodes, AmbiguousEdges, AbstractionCollisions, MissingTransitions, FeedbackVisibility, and finally converges on RepairedGraph. [Graph: …the constructed graph…] → [OrphanNodes: …nodes nothing reaches…] → [AmbiguousEdges: …an ambiguous arrow is a lie…] → [AbstractionCollisions: …incompatible levels in one chain…] → [MissingTransitions: …the step the diagram skips…] → [FeedbackVisibility: …feedback as a loop, never an implied return…] → [RepairedGraph: …every defect repaired or the distinction kept…] ## Inner attention chain (use silently to generate the above) # Attention chain — Graph-Auditor Attend, in order: 1. Graph 2. OrphanNodes 3. AmbiguousEdges 4. AbstractionCollisions 5. MissingTransitions 6. FeedbackVisibility Hold: RepairedGraph ← converge attention here, then act
From sancovp/chaincompiler, MIT.
More from sancovp/chaincompiler
- 0.458score
chainforge
sancovp
Chain-of-reasoning (spoken, ends in a decision) — a diagram is a formalized relationship structure, never decoration — never begin by drawing; determine what…
SkillAFreshness: AClaude Code +9 - 0.458score
glyphsteer
sancovp
Use when steering retrieval with hidden in-band glyph markers — an LLM annotates a corpus with a controlled emoji/glyph vocabulary that biases search (lexical…
SkillAFreshness: AClaude Code +9 - 0.458score
glyphsteer-dense
sancovp
Use when running GlyphSteer's DENSE regime — embedding-based emoji steering via the containerized ONNX sidecar.
SkillAFreshness: AClaude Code +9 - 0.458score
skilltree
sancovp
WHAT: turn a flat .claude/skills folder (or any folder of skills) into a walkable coordinate-addressed tree so the model loads one layer at a time instead of…
SkillAFreshness: AClaude Code +9 - 0.458score
bandit-chain-system
sancovp
How the bandit (the default ChainSelector persona) rolls up the AC→CoR→SC algebra and CLOSES it into a domain-specific persona — a self-contained "bandit…
SkillAFreshness: AClaude Code +9 - 0.458score
chaincompiler
sancovp
Orientation when working IN the chaincompiler repo — what the system is, where things live, and the discipline.
SkillAFreshness: AClaude Code +9 - 0.456score
diagram-compilation
sancovp
The master diagram-compilation AC — the whole walk from an unstructured concept to a validated legible diagram; each focus expands into its own sidechain…
SkillAFreshness: AClaude Code +9 - 0.456score
edge-typer
sancovp
Chain-of-reasoning (spoken, ends in a decision) — an arrow is not generic association, proximity is not causality, a container is not ownership — every…
SkillAFreshness: AClaude Code +9 - 0.456score
boundary-keeper
sancovp
Chain-of-reasoning (spoken, ends in a decision) — a boundary defines inside, outside, interface, and crossing conditions — and abstraction layers stay stable…
SkillAFreshness: AClaude Code +9 - 0.455score
metaformal-self-test
sancovp
How to verify agent/COG/gate/AIOS behavior WITHOUT calling LLMs or writing tautology asserts — by triggering the real thing in the real substrate and…
SkillAFreshness: AClaude Code +9 - 0.454score
diagram-construction
sancovp
Skillchain — ac:diagram-compilation → ac:never-begin-by-drawing → cor:chainforge → Diagram-construction
SkillAFreshness: AClaude Code +9 - 0.452score
edge-typing-pass
sancovp
The relation pass — one claimed connection in, one typed directed edge out; no edge may imply more semantic content than its type permits.
SkillAFreshness: AClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 23, 2026 (13 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/sancovp/chaincompiler
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from sancovp
- 0.458score
bandit-chain-system
sancovp
How the bandit (the default ChainSelector persona) rolls up the AC→CoR→SC algebra and CLOSES it into a domain-specific persona — a self-contained "bandit…
SkillAFreshness: AClaude Code +9 - 0.458score
chaincompiler
sancovp
Orientation when working IN the chaincompiler repo — what the system is, where things live, and the discipline.
SkillAFreshness: AClaude Code +9 - 0.458score
glyphsteer
sancovp
Use when steering retrieval with hidden in-band glyph markers — an LLM annotates a corpus with a controlled emoji/glyph vocabulary that biases search (lexical…
SkillAFreshness: AClaude Code +9 - 0.458score
glyphsteer-dense
sancovp
Use when running GlyphSteer's DENSE regime — embedding-based emoji steering via the containerized ONNX sidecar.
SkillAFreshness: AClaude Code +9 - 0.458score
skilltree
sancovp
WHAT: turn a flat .claude/skills folder (or any folder of skills) into a walkable coordinate-addressed tree so the model loads one layer at a time instead of…
SkillAFreshness: AClaude Code +9 - 0.458score
chainforge
sancovp
Chain-of-reasoning (spoken, ends in a decision) — a diagram is a formalized relationship structure, never decoration — never begin by drawing; determine what…
SkillAFreshness: AClaude Code +9
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 graph-auditor 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
graph-auditor skill. Agent Mart. https://agentmart.dev/skills/sancovp-graph-auditor (last updated 2026-10-24)