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/stefaniuk/loadoutWrote 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/commands/stefaniuk/loadout/util.doc-system)<a href="https://agentmods.dev/commands/stefaniuk/loadout/util.doc-system"><img src="https://agentmods.dev/badge/commands/stefaniuk/loadout/util.doc-system/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/commands/stefaniuk/loadout/util.doc-system"><img src="https://agentmods.dev/badge/commands/stefaniuk/loadout/util.doc-system.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 | $0.00022 | $0.01089 |
| Opus 5.5 | $0.00009 | $0.00436 |
| Sonnet 5.5 | $0.00004 | $0.00218 |
| Haiku 4.5 | $0.00002 | $0.00109 |
Grade A, and why
util.doc-system 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 1mo 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 — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
This prompt delegates to the system-documentation skill.
Run the system-documentation skill and resolve the request in this order:
- Detect combined produce-and-review requests. If the user asks to create,
establish, synchronise, update, refresh, regenerate, fix drift in, or
bring documentation up to date, and also asks to review, audit, check,
verify, validate, assess, or deep-review it, run a two-phase workflow:
- first resolve the producing phase through the rules below
- then run
audit - if
auditsurfaces resolvable drift, re-enter the producing phase to fix it, then re-run the documentation validation gates; stop after at most one corrective pass and report any drift that remains - report the two phases separately
- Resolve the step. If the user input names a step (
foundation,architecture,reference,explanation,how-to,tutorials,operations,audience-indexes,governance,all), use it. Otherwise default toall. - Resolve the mode for a single-phase request, or for the producing phase of
a combined request:
- an explicit mode wins
- an explicit legacy alias (
feature-sync,adr-sync,surface-sync,release-sync) maps through the skill's compatibility rules - a diff-based, PR-based, or changed-files-only review resolves to
pre-pr-review - a request to review, audit, check, verify, validate, assess, or
deep-review documentation quality, completeness, placement, or drift
resolves to
audit - a request to establish missing structure, create the canonical baseline,
or recover clearly missing anchors resolves to
establish - a request to synchronise, update, refresh, regenerate, or fix drift
resolves to
sync - when none of the above apply and no diff, PR, or explicit scope narrows
the run, default to
auditrather thansync; an unboundedsyncover a full repository is edit-heavy and should be requested explicitly
- Resolve scope. An explicit scope wins. Otherwise use changed files when diff or PR context is available; if not, use the repository root.
- When
allis resolved, run the ordered step sequence defined by the skill. Within each step, create or update only the evidenced outputs and the minimal landing pages or indexes allowed by that step and mode.syncmay legitimately produce zero edits when no drift is evidenced; in that case report "no changes; no drift detected" and proceed to validation. - If the user supplies an unrecognised or contradictory step, mode, alias, or scope, respond with an error that names the offending value and lists the accepted step and mode values. Do not start the run. Do not error solely because step, mode, or scope was omitted when the defaults above resolve it.
- Before writing, state the resolved step, mode, scope, and whether the run
is single-phase or combined.
auditandpre-pr-revieware non- destructive: they MUST NOT create, edit, rename, or delete files, and they do not need to execute validation gates. If you delegate either mode to a subagent, repeat the read-only contract in the subagent prompt. - For repositories with non-trivial documentation systems, prefer delegating inventory and drift analysis to a read-only exploration subagent before editing. Keep edits in the main agent so the planning context stays focused.
$ARGUMENTS
Examples:
establishreference sync docs/reference/auditpre-pr-review changed filessync then audit docs/reference/
Report using this skeleton; omit phase sections that did not run.
## Run report
### Resolved invocation
- step / mode / scope / phase (single or combined)
### Phase 1 - <mode>
- Canonical files created or updated
- Validation gates executed and their exit codes (documentation-scoped only)
- Unresolved drift by class and canonical target
### Phase 2 - audit (or pre-pr-review)
- Findings grouped by canonical target file, with severity and class
- Totals by severity and by class
- Overall verdict
- Recommended follow-ups (out of scope for this run)
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.
- 1mo ago First seen · 98 lines · 22 tokens per session scan A ba972327ad3a
util.doc-system is a command published in the GitHub repository stefaniuk/loadout (1 stars, last pushed 13d ago), licensed MIT. It adds 22 tokens to every session and 1,089 once invoked, about $0.0001 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 2026-09-12.
Other commands, from other repositories
sddp-implement-qc-loop
Command description: Run implement and QC in a continuous loop. Argument hint: [optional: feature directory or branch name] Command category: orchestration Prerequisites: spec, plan, tasks.
sddp-plan
Command description: Create an implementation plan from the current feature specification. Argument hint: [optional: planning constraints or focus areas] Command category: feature-delivery Prerequisites: spec.
sddp-prd
Command description: Create or refine the canonical product document. Argument hint: [rough product idea, users, domain, or market opportunity] Command category: project-bootstrap Prerequisites: none.
sddp-qc
Command description: Run quality control against the implemented feature. Argument hint: [optional: testing focus such as unit tests, security audit, requirements sync] Command category: feature-delivery Prerequisites: spec, plan, tasks, implementation:complete.
sddp-specify
Command description: Create a feature specification from a feature description. Argument hint: [feature description] Command category: feature-delivery Prerequisites: project-instructions.
sddp-amend
Command description: Propagate a bootstrap change across canonical project artifacts and the project plan. Argument hint: [project-level change to propagate across bootstrap artifacts] Command category: project-bootstrap Prerequisites: project-instructions, product-document, technical-context, project-plan.