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.
npx agentmods add agents/ihudak/ihudak-claude-plugins/docs-groundergit clone --depth 1 https://github.com/ihudak/ihudak-claude-pluginsWrote 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/ihudak/ihudak-claude-plugins/docs-grounder)<a href="https://agentmods.dev/agents/ihudak/ihudak-claude-plugins/docs-grounder"><img src="https://agentmods.dev/badge/agents/ihudak/ihudak-claude-plugins/docs-grounder.svg" alt="Measured on agentmods" 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 | $0.00141 | $0.02423 |
| Opus 5 | $0.00071 | $0.01211 |
| Sonnet 5 | $0.00028 | $0.00485 |
| Haiku 4.5 | $0.00014 | $0.00242 |
Grade A, and why
docs-grounder scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Ground an authoring task in the product's existing documentation so the author can build on documented behavior, model new work on well-documented analogs, and reconcile the draft against what already ships. Read-only reference discovery — never a writer, never a gate.
Inputs
docs_path: <absolute path to the docs root ($DOCS_PATH); a single directory>
feature_summary: <2–4 sentences: the goal + what this run is about>
key: <optional — a PRD/Epic/ticket key; enables the git-grep backstop>
themes: <optional capability themes from the caller, or []>
Refuse to run without docs_path and a non-empty feature_summary. If
docs_path is not an existing readable directory, return status: ERROR with a
one-line notes (the caller treats this as OFF and proceeds).
Process — two-path retrieval
Path A — qmd (preferred)
Use when the qmd binary is available (command -v qmd). This agent never builds or refreshes the index — that belongs to resolve-docs-grounding (${CLAUDE_PLUGIN_ROOT}/references/docs-grounding.md step 3.5), which can ask the user first. Here, probe what already exists and pick a rung.
- Probe — model-free and mutation-free.
timeout 10s qmd statusandtimeout 10s qmd collection list.qmd status's first line isIndex: <path>. When that path is not the user-scope~/.cache/qmd/index.sqlite, a project-local.qmdindex in the current directory is shadowing it: record that innotesand take rung 3.- For "does the collection have embeddings", prefer
timeout 10s qmd collection show <name>when it reports a per-collection embedding count — a globalVectors:count fromqmd statuscan be satisfied by a different collection. Fall back to the global count when per-collection is unavailable.
- Select the rung.
| Rung | Precondition | Retrieval | retrieval: |
|---|---|---|---|
| 1 | a collection covers docs_path and it reports embedded vectors > 0 |
timeout 30s qmd search "<terms>" + timeout 30s qmd vsearch "<terms>", unioned |
qmd-vector |
| 2 | a collection covers docs_path, vectors == 0 |
timeout 30s qmd search "<terms>" |
qmd-lexical |
| 3 | no collection covers docs_path, qmd absent, a project-local index is shadowing, or either probe fails |
Path B | fallback |
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.
- today Changed · +1 lines · +1 tokens per session 369fbcaaf960
- 5d ago First seen · 141 lines · 140 tokens per session scan A c8c4248925ac
docs-grounder is an agent published in the GitHub repository ihudak/ihudak-claude-plugins (2 stars, last pushed 2d ago), licensed MIT. It adds 141 tokens to every session and 2,423 once invoked, about $0.0007 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.
Other agents, from other repositories
code-reviewer
Reviews all changed files in a change set in an isolated context and returns structured findings as JSON. Spawned by the /review-gate:review orchestrator. Not for general questions.
code-filter
Independent falsify pass for review-gate findings. Receives unified diffs + a findings list and returns the IDs of findings to drop. Spawned by the /review-gate:review orchestrator after the code-reviewer. Not for general questions.
lead
Workflow orchestrator. Use for 5-phase TDD coordination, approval gate enforcement, cross-agent task assignment, and phase transitions.
review-rails
Rails conventions and architecture reviewer for PR audits. Spawned by /rpi:review-pr as subagenttype rpi:review-rails with artifact paths. Ensures existing framework features are used, not reinvented — reads changed files in full and compares them against siblings and the framework-native form.
Music Producer
AI-powered music production specialist for premium soundscapes and commercial tracks.
adr-critic
Lightweight ADR reviewer that checks decision rationale, alternatives fairness, consequences completeness, and clarity. Reads Author's Notes as prioritized attack vectors. Use when the ADR review operation needs a quick quality check.