agentmods.dev

re-anchor

re-anchor is a skill for Claude Code from goldenwo/claude-state-drift. It costs 108 tokens per session (1,173 once invoked), scanned A, original, MIT.

A workflow check that rereads the project's main objective and compares it with recent work, version-control history, and uncommitted changes.

In plain words
What is it for?
Use it after a long work session, when the task seems to be drifting, after resuming paused work, or when you want to confirm the work is still on track.
Why use it?
It helps catch scope drift, such as spending time on extra refactoring instead of the requested task. It also provides a final direction check before work is declared complete.

Skill for Claude Code

Written for Claude Code: allowed-tools in frontmatter.

Part of the claude-state-drift plugin — 3 skills, 6 commands, 4 hooks shipped together

not rated 1repo 24d ago A scan 108 tokens original MIT

Good fit Use it after a long work session, when the task seems to be drifting, after resuming paused work, or when you want to confirm the work is still on track.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/goldenwo/claude-state-drift/re-anchor
View source ↗ goldenwo/claude-state-drift
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.

Any agent
npx skills add goldenwo/claude-state-drift --skill re-anchor
Clone the repo
git clone --depth 1 https://github.com/goldenwo/claude-state-drift

Made for: Claude Code.

Or install claude-state-drift, the plugin that ships this one along with the rest of its 3 skills, 6 commands, 4 hooks.

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 re-anchor

README.md
[![agentmods](https://agentmods.dev/badge/skills/goldenwo/claude-state-drift/re-anchor/github.svg)](https://agentmods.dev/skills/goldenwo/claude-state-drift/re-anchor)
Your own site
<a href="https://agentmods.dev/skills/goldenwo/claude-state-drift/re-anchor"><img src="https://agentmods.dev/badge/skills/goldenwo/claude-state-drift/re-anchor/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for re-anchor

Your own site · 80×15
<a href="https://agentmods.dev/skills/goldenwo/claude-state-drift/re-anchor"><img src="https://agentmods.dev/badge/skills/goldenwo/claude-state-drift/re-anchor.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 108 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,173 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00108 $0.01173
Opus 5 $0.00054 $0.00587
Sonnet 5 $0.00022 $0.00235
Haiku 4.5 $0.00011 $0.00117

Measured 11d ago against content hash ec5329db8f4a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

re-anchor 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 11d 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.

skills/re-anchor/SKILL.md · 82 lines

How it starts

The opening of the file, as written. The whole thing — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.

re-anchor

Pause, re-read the master objective, audit the current session's recent work for alignment, report. Inspired by Reflexion's "re-pass the original task each iteration" pattern.

When to invoke

  • After ~30+ minutes of intensive work without checking back to the objective.
  • When noticing scope drift mid-session (extra refactors, shaving yaks, exploratory rabbit holes).
  • Before declaring a task complete — sanity-check that what was built is what was asked for.
  • On user request: "re-anchor", "check we're on track", "are we still on the right path", "sanity check direction".
  • After resuming a session that was paused for hours/days.

This skill does NOT modify state.json — that's update-state's job. Its only write is one verdict line appended to the state-history.jsonl audit log (step 4).

Procedure

  1. Re-read the master objective. Read .claude/state.json and extract objective, version, current_focus, and the IDs of in_progress deliverables.

  2. Inspect recent work. Recent assistant turns (the last 10-20 messages of the session), git log --oneline -10, git status for uncommitted changes. Identify what was actually worked on.

  3. Compare and assess. Produce a short audit with this structure:

    === RE-ANCHOR AUDIT ===
Original objective: <verbatim from state.json>
Current version:    <state.version>
State.json current_focus: <verbatim>
In-progress deliverables: <list of titles>

Recent work (this session): <one-paragraph summary of what was actually done>

Drift assessment: <on-track | mild | significant>
- On-track: recent work directly advances in-progress deliverables AND current_focus
- Mild: recent work is adjacent (e.g., refactor that supports the deliverable) but not on the critical path
- Significant: recent work is unrelated to the objective; new scope discovered or attention captured elsewhere

Recommendations:
- <one to three concrete next moves>

  4. Log the verdict. Append the drift verdict to the project's append-only audit log so accumulated logs can answer "how often is drift actually caught?":

Read the full file on GitHub · 82 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. 11d ago First seen · 82 lines · 108 tokens per session scan A ec5329db8f4a

Subscribe to this mod's changes

re-anchor is a skill published in the GitHub repository goldenwo/claude-state-drift (1 stars, last pushed 24d ago), licensed MIT. It adds 108 tokens to every session and 1,173 once invoked, about $0.0005 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 2026-08-31.

Claude CodePlanningclaude-code-plugindrift-mitigationstate-tracking
Related

Other skills, from other repositories

create-site

Creates a new Power Pages code site (SPA) using React, Angular, Vue, or Astro. Guides through the full process from initial concept to deployed site: requirements discovery, scaffolding, component planning, design, implementation, validation, and deployment. Use when the user wants to create, build, or scaffold a new…

microsoft/power-platform-skills · 73 tokens

google-ads-audit

Google Ads account audit and business context setup. Use for account-health audits and business-context setup. Trigger on "audit my ads", "ads audit", "set up my ads", "onboard", "account overview", "how's my account", "ads health check", "what should I fix in my ads", or when the user is new to NotFair and hasn't run…

nowork-studio/notfair-plugin · 86 tokens

review

5-pass structured code review — correctness, security, performance, readability, consistency.

SethGammon/Citadel · 17 tokens

alive:system-upgrade

Upgrade ALIVE to the current version. Handles v1/v2/v3.x source states, multi-surface aware (alive-mcp / Hermes / Codex), retroactive version detection, partial-failure resume, dry-run previews, and rollback inspection.

alivecontext/alive · 57 tokens

extract-resume

Parse a resume's uploaded PDF into structured JSON (basics, experience, projects, skills, education) and save it to the editor.

suxrobGM/jobpilot · 32 tokens

ensure-pipelines-host

Ensures the tenant has a usable Power Platform Pipelines host environment before any pipeline operation runs. Detects host state via the same resolution order as the Power Apps UI (org-db setting → BAP env metadata → default-custom-host setting); if any existing host (Platform or Custom) is found, uses it. If no host…

microsoft/power-platform-skills · 222 tokens