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 skills add anshss/shiploop --skill deep-research-tieredgit clone --depth 1 https://github.com/anshss/shiploopWrote 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/skills/anshss/shiploop/deep-research-tiered)<a href="https://agentmods.dev/skills/anshss/shiploop/deep-research-tiered"><img src="https://agentmods.dev/badge/skills/anshss/shiploop/deep-research-tiered/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/skills/anshss/shiploop/deep-research-tiered"><img src="https://agentmods.dev/badge/skills/anshss/shiploop/deep-research-tiered.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 | $0.00000 | $0.00516 |
| Opus 5 | $0.00000 | $0.00258 |
| Sonnet 5 | $0.00000 | $0.00103 |
| Haiku 4.5 | $0.00000 | $0.00052 |
Grade A, and why
deep-research-tiered 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 10d 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.
What it actually says
deep-research-tiered — model-tiered deep-research (workspace override)
Shiploop workspaces ship a model-tiered override of the built-in deep-research workflow. Prefer
it over the built-in in this workspace:
Workflow({name: 'deep-research-tiered', args: '<question>'}) // cheap-tier defaults
Workflow({name: 'deep-research-tiered', args: {question: '<q>', models: { // brain overrides
scope: 'sonnet', search: 'sonnet', fetch: 'haiku',
verify: 'sonnet', synthesize: 'opus',
}}})
Same output shape as the built-in. The five stages (Scope → Search → Fetch → Verify → Synthesize)
each accept an explicit model pin via args.models. Null-semantics contract:
- absent OR explicit
nullfor a stage → the tiered default - literal string
"inherit"→ no model pinned (session model handles the stage) - any other string → pin that model
The tiered defaults keep a brainless invocation cheap: scope=sonnet, search=sonnet,
fetch=haiku (with effort: 'low'), verify=sonnet, synthesize=inherit. Only synthesis
inherits the session model by default — that's the one stage where the frontier tier pays.
Naming and precedence
The .claude/workflows/deep-research.js file is renamed at meta.name to deep-research-tiered
so it can't collide with the built-in by name (whether a same-named workspace copy shadows the
built-in is undocumented in the Workflow tool spec — this fallback is robust either way).
Invoking Workflow({name: 'deep-research-tiered', ...}) unambiguously routes to the workspace
copy in every session.
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.
- 10d ago First seen · 37 lines · 0 tokens per session scan A 2d5268bbde7f
deep-research-tiered is a skill published in the GitHub repository anshss/shiploop (7 stars, last pushed today), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 516 tokens. 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 skills, from other repositories
design
Create a doc-as-code design package from a PRD or SPEC. Conditionally generates C4 diagrams (Context/Container/Component), sequence diagrams, ER diagram + Data Dictionary, OpenAPI 3.0, AsyncAPI 3.0, ADRs, domain glossary, state diagrams, and deployment view as Mermaid-rendered Markdown files. Use when PM mentions…
design-corpus
EXPERIMENTAL (Claude Code only). Apply a SPEC increment to a single LIVING architecture corpus under docs/architecture/ instead of a per-SPEC silo package — treats docs as event-sourcing (SPEC = commit, corpus = working tree), so C4 Context/Container, glossary and the data model stay system-wide and never drift across…
review-pr
A pull-request review skill for evaluating proposed code changes and posting a quality verdict. A pull request is a request to merge changes into a shared codebase.
tasks
A task-breakdown tool that turns a plan, specification, feature brief, bug report, technical-debt item, or chore into small TASK-NNN work items. Technical debt means postponed cleanup or design work in a codebase.
continue
Autonomous work — find and execute ready tasks.
reconcile-docs
Best-effort reconciliation of the living architecture corpus against the code: the model reads docs/architecture/ plus the code and lists divergences with coordinates. Advisory only — an opinion of the model, never a verified verdict, no gates and no stamps. Nothing is applied automatically: corpus edits are a…