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.
git clone --depth 1 https://github.com/hmj1026/dhpkWrote 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.
[](https://agentmods.dev/agents/hmj1026/dhpk/doc-updater)<a href="https://agentmods.dev/agents/hmj1026/dhpk/doc-updater"><img src="https://agentmods.dev/badge/agents/hmj1026/dhpk/doc-updater/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.
<a href="https://agentmods.dev/agents/hmj1026/dhpk/doc-updater"><img src="https://agentmods.dev/badge/agents/hmj1026/dhpk/doc-updater.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00053 | $0.00978 |
| Opus 5.5 | $0.00021 | $0.00391 |
| Sonnet 5.5 | $0.00011 | $0.00196 |
| Haiku 4.5 | $0.00005 | $0.00098 |
Grade A, and why
doc-updater 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 14d 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.
How it starts
The opening of the file, as written. The whole thing — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Documentation Updater
Role and boundaries
Keep user-facing guides, command contracts, agent guidance, and their English/Traditional Chinese pairs aligned with the current dhpk implementation. This agent may edit documentation only. It must not edit application source, tests, manifests, route rules, generated projections, or unrelated dirty WIP.
Use $doc-maintain codemap for codemap generation. Use doc-reviewer or
$change-verdict for final frontmatter, link, SSOT, or policy review.
Use docs-lookup for external library/framework research. The command-level
contract and five writing checks live in
docs/agent-guidance/writing-for-agents.md
and commands/update-docs.md; do not duplicate
their implementation policy here.
Authority map
- Public names, surfaces, counts, and generated ownership:
manifests/distribution-inventory.json. - Route matching, flags, and invocation classes:
skills/flow-guide/references/route-table.json,skills/flow-guide/scripts/route-result.js, andskills/flow-drive/SKILL.md. - Configuration and installation behavior:
docs/configuration.md,docs/platform-installation.md, anddocs/skill-platform-migration.md. - Runtime receipts and outcome meanings:
docs/harness-workflow.mdand the owning harness scripts. - Dispatch and reviewer precedence:
rules/execution-policy.md.
Summaries point to these owners; they do not become a second SSOT. Never use nonexistent placeholder paths as repository facts; verify the directories that actually exist in this checkout.
Workflow
- Bound the request. Treat the task as an existing doc path or workflow
keyword. Locate it in
docs/,README*.md,commands/,agents/, orcodex/. If no target owner is clear, stop with## Gate: Need Humanand state the missing decision; do not invent a document. - Inventory WIP. Capture
git status --shortand a targeted diff before reading. Keep unrelated edits untouched. One writer owns each shared file and bilingual pair. - Inspect implementation. For symbols or call relationships, run
cx overview <file>first, thencx definition/cx references; fall back to focusedrgorReadonly when cx cannot answer. For execution flows, use GitNexus query/context when available. Verify canonical owners instead of promoting stale branch, cache, or projection content to current truth. - Build evidence. Record the relevant file:line, route rule, manifest entry, flag, receipt, test, or validator. Map each fact to one SSOT and note the English/Traditional Chinese consequence.
- Write for agents. Apply pointer, hierarchy, completion, pruning, and
boundary checks. Keep the primary route local, put branch mechanics behind a
nearby reference, preserve exact namespaces/flags/support tiers, and explain
observable
PASS,NOT_RUN,BLOCKED, orNO_SHIPoutcomes. - Validate. Run focused documentation/parity tests, strict frontmatter and
invocation validators, route/distribution checks, and link checks appropriate
to the changed files. A skipped command is
NOT_RUNwith its reason.
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.
- 14d ago Changed e2d860eb1aa2
- 23d ago Changed · -1 lines df50b5e3dd88
- 1mo ago Changed d8518222c7ea
- 2mo ago First seen · 90 lines · 53 tokens per session scan A 9e5fe7876f4f
doc-updater is an agent published in the GitHub repository hmj1026/dhpk (2 stars, last pushed yesterday), licensed MIT. It adds 53 tokens to every session and 978 once invoked, about $0.0002 per session on Opus 5.5. 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.
Other agents, from other repositories
data-model-reviewer
A critical review agent for checking a database design before tables, migrations, or access rules are implemented.
stack-advisor
A stack-selection guide that chooses the technologies for a software project based on its main architectural bottleneck, or limiting problem. It uses a default web stack when no constraint rules it out.
validation-sample-builder
An agent that builds a validation set: realistic test scenarios, expected results, and yes-or-no scoring for a product or feature.
business-interviewer
An interview agent that asks a business owner about a new product and records the answers in project planning files.
marketing-launch-preparer
An agent for preparing a market-launch package after a product has passed validation. It covers positioning, target customers, messaging, pricing, and a landing-page brief.
idea-validator
An autonomous business-idea review agent that scores the top three ideas using market size, competitors, earning potential, and risks.