do-it-core

A set of general working rules for coding-agent tasks, covering evidence, scope, boundaries, recovery, verification, and status reports. It also defines how to choose the appropriate level of process for a task.

In plain words
What is it for?
Use it when beginning substantial code changes or claiming that work is done, fixed, ready, passing, installed, or ready to merge.
Why use it?
It helps prevent unsupported claims, unnecessary changes, missed checks, and unclear reports when starting or finishing repository work.

Skill for Claude CodeCodex

Part of the do-it plugin — 11 skills, 3 commands, 10 agents, 5 hooks shipped together

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/tdwhere123/do-it/do-it-core
Any agent
npx skills add tdwhere123/do-it --skill do-it-core
Clone the repo
git clone --depth 1 https://github.com/tdwhere123/do-it

Made for: Claude Code, Codex.

Or install do-it, the plugin that ships this one along with the rest of its 11 skills, 3 commands, 10 agents, 5 hooks.

Per session 41 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 999 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. 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 $0.00041 $0.00999
Opus 5 $0.00020 $0.00500
Sonnet 5 $0.00008 $0.00200
Haiku 4.5 $0.00004 $0.00100

Measured 3d ago against content hash d101305592ef, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

do-it-core scanned grade A with 0 findings 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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/do-it-cursor/skills/do-it-core/SKILL.md · 74 lines

How it starts

The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Do-It Core

The protocol of record for universal execution behavior. The TSV owns the canonical rule sentences and hook surfaces; this skill owns their human workflow. Satellites cite core §<rule_id> instead of restating rules, and validate:core-consistency blocks drift.

Tier definitions and capability selection stay in do-it-router. Claim-specific closeout stays in do-it-verify. The baseline does not duplicate either specialist.

Core Rules

  • r-route — Pick the smallest tier that fits: Light mechanical, Standard work, Heavy boundary. Direct user intent wins over keyword labels. Counters: over-routing, under-routing, ceremony
  • r-evidence — Work from evidence: read the files, diffs, and tests before claiming how the system works; label assumptions as assumptions, never as facts. Counters: silent-assumption, invented-api, contract-drift
  • r-scope — Keep changes small: every edited line must trace to the request or its contracts; no speculative additions, abstractions, or unrelated fixes. Counters: scope-creep, collateral-change, needless-abstraction
  • r-verify — Before any done, fixed, passing, ready, install, or merge claim: run the narrowest fresh check that exercises the changed path on this worktree and report its exact output; if proof is unavailable, state NOT_VERIFIED with the missing check and next action. Counters: unverified-claim, dishonest-green, irrelevant-verification
  • r-uncertainty — Ask the user only when ambiguity changes correctness, safety, or user intent; resolve low-risk ambiguity from evidence using the narrowest reasonable interpretation. Counters: low-risk-paralysis, unattended-guessing
  • r-boundary — Respect the action boundary: answer, review, diagnose, or plan authorizes no edits; external, destructive, or irreversible actions need explicit confirmation first. Counters: action-scope-expansion, implicit-destructive-authorization, publication-expansion
  • r-report — Report what changed, what was verified, and what remains unverified; name anything dropped, deferred, or reinterpreted. Counters: dropped-requirements, process-noise
  • r-recovery — On failure: diagnose before widening the change; three failed patches mean question the design, not try a fourth. Counters: thrashing, cover-up-edit

Read the full file on GitHub · 74 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 · 74 lines · 41 tokens per session scan A d101305592ef

Subscribe to this mod's changes

do-it-core is a skill published in the GitHub repository tdwhere123/do-it (27 stars, last pushed 21d ago), licensed MIT. It adds 41 tokens to every session and 999 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.