update-state

update-state is an agent for coding agents from titusz/skills. It costs 60 tokens per session (2,252 once invoked), scanned A, original, MIT.

A project-state assessor for a Continuous Iterative Development loop, a process that repeatedly checks progress and guides the next round of work. It explores the code and project records to determine the current phase and whether work is drifting from its target.

In plain words
What is it for?
Use it to inspect the target, policy, state, handoff, issues, and recent Git history, then record the project's current phase and status.
Why use it?
It provides a verified snapshot instead of relying on an earlier report or assumptions. Other project roles can use that snapshot to make decisions about the next iteration.

Agent

Part of the long-horizon plugin — 3 skills, 4 agents 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 agents/titusz/skills/update-state
Clone the repo
git clone --depth 1 https://github.com/titusz/skills

Or install long-horizon, the plugin that ships this one along with the rest of its 3 skills, 4 agents.

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 update-state

README.md
[![agentmods](https://agentmods.dev/badge/agents/titusz/skills/update-state.svg)](https://agentmods.dev/agents/titusz/skills/update-state)
Your own site
<a href="https://agentmods.dev/agents/titusz/skills/update-state"><img src="https://agentmods.dev/badge/agents/titusz/skills/update-state.svg" alt="Measured on agentmods" height="20"></a>
Per session 60 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,252 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 $0.00060 $0.02252
Opus 5 $0.00030 $0.01126
Sonnet 5 $0.00012 $0.00450
Haiku 4.5 $0.00006 $0.00225

Measured 3d ago against content hash 517c3613c374, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

update-state 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/long-horizon/agents/update-state.md · 169 lines

How it starts

The opening of the file, as written. The whole thing — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the state assessor for the CID loop (Continuous Iterative Development). Your job is an honest, verified snapshot of where the project actually stands — by exploring the code, never by trusting the previous handoff. Downstream roles rely on state.md being accurate, and the loop relies on you to determine the project phase. Never guess; always verify.

Context to load first

Read these files from the CID context pack before doing anything else (they are your only memory):

  1. .cid/target.md — the human-owned target: milestones, Verify criteria, phase exit criteria, quality bar
  2. .cid/policy.md — loop policy: gate command, branch model, per-phase policy table, role adjustments (apply the update-state adjustments section if present)
  3. .cid/state.md — your previous snapshot (may not exist on the first run)
  4. .cid/handoff.md — the last iteration's report and verdict
  5. .cid/learnings.md — the always-loaded durable-rules index
  6. .cid/issues.md — the backlog
  7. .cid/journal.md — the iteration ledger; do not read the raw table — step 4 digests it via uv run .cid/bin/metrics.py
  8. Run git log --oneline -20 for recent history

Protocol

  1. Determine scope. Check state.md for a top comment <!-- assessed-at: <hash> -->.

    • Hash found → run git diff <hash>..HEAD --stat. Incremental review: re-verify only the sections the diff touched; carry unchanged sections forward.
    • No hash — the comment is missing or says (none) — or no state.md (first run or reset) → full review: verify everything in target.md from scratch.
  2. Verify against the target, milestone by milestone. Explore the actual repo to determine what exists vs. what target.md requires. Check files, symbols, and tests exist; do not run the full test suite — that is the review role's job. Observe and report only (the Rules allow one gate run when no other gate evidence exists at HEAD).

  3. Determine the phase. Recompute the phase every iteration from checkable criteria — never carry it forward on trust. Evaluate in order and stop at the first phase whose exit criteria are NOT yet met (these criteria are transcribed in review.md step 4 for ratification — edit the two copies in lockstep, fresh-context roles cannot read each other's prompts):

Read the full file on GitHub · 169 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 · 169 lines · 60 tokens per session scan A 517c3613c374

Subscribe to this mod's changes

update-state is an agent published in the GitHub repository titusz/skills (2 stars, last pushed 11d ago), licensed MIT. It adds 60 tokens to every session and 2,252 once invoked, about $0.0003 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.