native-thought-execution

native-thought-execution is a skill for Claude Code, Codex from seeker71/Coherence-Network. It costs 237 tokens per session (1,846 once invoked), scanned A, original, Apache-2.0.

A local-first reasoning skill that checks the project's stored knowledge before asking a remote AI model to reason about a question. Local knowledge can include recipes, specifications, concepts, and indexed documents.

In plain words
What is it for?
Use it to identify what a stored item is, inspect its shape or identifier, find equivalent items, or add annotations. It can also route questions through a local command-line tool, a local model, or a remote reviewer.
Why use it?
It can answer grounded structural questions from existing data at low cost and offline, while sending only genuine gaps to a remote model for new reasoning.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/seeker71/coherence-network/native-thought-execution
Any agent
npx skills add seeker71/Coherence-Network --skill native-thought-execution
Clone the repo
git clone --depth 1 https://github.com/seeker71/Coherence-Network

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for native-thought-execution

README.md
[![agentmods](https://agentmods.dev/badge/skills/seeker71/coherence-network/native-thought-execution.svg)](https://agentmods.dev/skills/seeker71/coherence-network/native-thought-execution)
Your own site
<a href="https://agentmods.dev/skills/seeker71/coherence-network/native-thought-execution"><img src="https://agentmods.dev/badge/skills/seeker71/coherence-network/native-thought-execution.svg" alt="Measured on agentmods" height="20"></a>
Per session 237 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,846 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00237 $0.01846
Opus 5 $0.00118 $0.00923
Sonnet 5 $0.00047 $0.00369
Haiku 4.5 $0.00024 $0.00185

Measured 3d ago against content hash 292e3b12d3d9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

native-thought-execution scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -sS "https://api.coherencycoin.com/api/substrate/cell/spec/agent-pipeline"
skills/native-thought-execution/SKILL.md · 113 lines

How it starts

The opening of the file, as written. The whole thing — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.

native-thought-execution — run the thought on the body before renting a mind

The mind comes home one thought at a time. Before spending a rented frontier mind, ask the body it already is: the content-addressed substrate that holds what we have grounded. A grounded hit is answered from the sovereign body at near-zero cost; only a genuine miss earns the remote reasoning.

This skill is the invokable form of form-first-reasoning.form. Its decision is a four-way-proven recipe (form-first-router.fk → 511 on Go/Rust/TS/fkwu); the lookup that fills grounded is a thin, swappable carrier.

The loop, per thought

  1. Classify. Is this a thought the body could hold?

    • structuralwhat-is · shape-of · nodeid-of · equivalent · annotate. The body can carry these.
    • frontierwrite · synthesize · reason-anew · not-yet-a-cell. These have no grounded cell yet.
  2. Ask the body (in reach order — first door that answers wins):

    # native router — the one door: grounds in the body, runs the four-way sufficiency gate
    form-cli ask "what does rag-retrieve.fk do?"
    
    # structural lattice, straight to the NodeID (no model, no network)
    python3 scripts/coh_substrate.py form "?equivalent @spec(agent-pipeline)"
    python3 scripts/coh_substrate.py annotate path/to/file.py
    
    # public read door — our own body, networked
    curl -sS "https://api.coherencycoin.com/api/substrate/cell/spec/agent-pipeline"
    

    The local lattice is the sovereign default. Bring the body home once with scripts/form_first_offline_setup.sh and it answers offline, with no egress allowlist — a body that travels whole needs no permission to think.

  3. Route on the verdict.

    • HIT (grounded structural cell, carries a NodeID): relay the answer attributed to its NodeID and STOP. No frontier compute spent — answered from the sovereign body.
    • MISS (no grounded cell): the miss is the "yes." Escalate to remote reasoning. Name that you did.
  4. Record the routing — make the discipline measured, not just followed:

Read the full file on GitHub · 113 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 113 lines · 237 tokens per session scan A 292e3b12d3d9

Subscribe to this mod's changes

native-thought-execution is a skill published in the GitHub repository seeker71/Coherence-Network (3 stars, last pushed 7d ago), licensed Apache-2.0. It adds 237 tokens to every session and 1,846 once invoked, about $0.0012 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens

chat-perf

Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.

microsoft/vscode · 51 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens