Skip to content
Agentmart
Submit a tool

operating-model

Use when shaping agent prose or making code changes: answer design, lazy-correct engineering, and the AI-tell editing catalog.

Skill by Exploitacious

IndexedMIT
Write a review

Chalk Labs

Need help deploying operating-model?

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.

Book a call

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/exploitacious-operating-model

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.

We use your email only for this newsletter. Privacy

About

Use when shaping agent prose or making code changes: answer design, lazy-correct engineering, and the AI-tell editing catalog.

Use cases: design, writing, agent-building

No 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/operating-model

What's inside

Skill name
operating-model
Description
Use when shaping agent prose or making code changes: answer design, lazy-correct engineering, and the AI-tell editing catalog.
License
MIT
Instructions
49 lines in SKILL.md · view on GitHub

Files (3)

  • SKILL.md
  • adhd-output.md · references/
  • patterns.md · references/

SKILL.md (first lines)

# Operating model

Detail for the active writing and engineering contract. Where the harness supplies a core (`CONTEXT/core.md`, its Writing section), that file owns the standing rules; where it is absent, follow the active instructions. This skill owns answer-shaping technique and the engineering ladder; the active contract owns the standing rules.

## Talk terse, the detail

Useful rewrites: "use" for "utilize", "fix" for "implement a solution for", "big" for "extensive". Preserve identifiers, error strings, counts, file paths, and API shapes while cutting filler. The active contract owns register exceptions.

## Make the answer easy to act on

- Number steps the reader must execute, bound each step, and make the first small enough to start now.
- During ongoing work, show what changed and what the next step will resolve. After an interruption, restore the relevant state without replaying the plan.
- Group related items and rank by relevance, about five per visible group. Presentation limits never reduce required research or content. When the operator is choosing priorities or what to work next, keep the whole menu with neutral facts (deadline, size, owner, state), then give your recommendation after it, marked as yours.
- Keep tangents in notes; surface dependencies or risks that affect the task. When reader action is required, identify the one decision or action that unblocks it. Avoid inventing homework after completion.
- Explain an error as what failed, the known cause, and the next diagnostic or fix. Give useful time estimates in concrete units with their assumptions. Preserve uncertainty, empathy, and essential safety or decision details.

Source and integration notes: [references/adhd-output.md](references/adhd-output.md).

## Build lazy-correct

Understand the task and real code flow first. Stop at the first adequate rung:

1. Is the need speculative? Skip it and say why in one line.
2. Reuse a helper, type, utility, or pattern already in the codebase.
3. Use the standard library when it covers the need.
4. Prefer a native platform feature over a dependency.
5. Use an installed dependency; do not add one for work a few lines cover.
6. Use one line when one line does the job.
7. Otherwise write the minimum working code.

Find every caller before changing a function and fix the shared cause once. Avoid unrequested abstractions: a one-implementation interface, a one-product factory, or configuration for an unchanging value. Prefer deletion and boring code over cleverness. Keep trust-boundary validation, protection against data loss, security controls, accessibility basics, and every requested behavior.

Tag a deliberate shortcut `shortcut:` with its ceiling and upgrade path, for example `# shortcut: global lock, per-account locks if throughput matters`. The literal tag exists so one search lists every deliberate shortcut when someone later asks what debt the code carries.

For the active test contract (`CONTEXT/core.md`, its landing section, and `CONTEXT/how-i-work.md` P9, "Tests scale with the work"), choose the smallest check that fails on broken behavior: an assert-based self-check or one focused test for a branch, loop, parser, money path, or security path. Trivial one-liners need no new test.

## No AI tells

Load [references/patterns.md](references/patterns.md) when editing anything longer than a chat reply. It owns the detailed editing catalog. For writing on the operator's behalf, use the active voice instructions and `CONTEXT/voice.md` when present (the `operator-voice` skill loads it); otherwise use plain professional prose.

## Steward the shared tooling

`stewardship` owns defect triage and source fixes for the shared tools. Use `meta-skill-creator` to capture reusable guidance from corrections and recurring friction. Keep routine success quiet; an observation is a candidate, not a standing order.

## Every lane too

Scope and reporting follow the active delegation contract (`CONTEXT/core.md`, its lanes section). The same writing and engineering techniques apply inside a lane.

From Exploitacious/OPS, MIT.

More from Exploitacious/OPS

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Oct 19, 2026 (12 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from Exploitacious

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 operating-model 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.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying operating-model?

CallGet help
Cite this page

operating-model skill. Agent Mart. https://agentmart.dev/skills/exploitacious-operating-model (last updated 2026-10-24)