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 skills/obsidian-owl/specwright/sw-researchnpx skills add Obsidian-Owl/specwright --skill sw-researchgit clone --depth 1 https://github.com/Obsidian-Owl/specwrightWrote 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/obsidian-owl/specwright/sw-research)<a href="https://agentmods.dev/skills/obsidian-owl/specwright/sw-research"><img src="https://agentmods.dev/badge/skills/obsidian-owl/specwright/sw-research.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 5.1 | $0.00029 | $0.00742 |
| Opus 5 | $0.00015 | $0.00371 |
| Sonnet 5 | $0.00006 | $0.00148 |
| Haiku 4.5 | $0.00003 | $0.00074 |
Grade A, and why
sw-research 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 6d 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 — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Specwright Research
Goal
Produce validated, referenced research briefs as input to the design phase. Focus is outward — external docs, APIs, SDKs, industry patterns. Output is facts and evidence, never design opinions. Operates fully autonomously — briefs are consumed by sw-design (which has its own gate).
Inputs
- Research topic(s) from argument or conversation context
{projectArtifactsRoot}/research/— existing briefs (for deepening or refresh){projectArtifactsRoot}/CHARTER.md— technology vision (for relevance filtering)
Outputs
{projectArtifactsRoot}/research/{topic-id}-{YYYYMMDD}.mdperprotocols/research.md
Constraints
Stage boundary (LOW freedom):
Reads and researches. NEVER writes code, branches, mutates workflow state,
or produces design artifacts. No currentWork, no lock. Can run anytime.
It is not a core workflow stage and never claims top-level work ownership.
Triage (MEDIUM freedom):
- Break request into 1-5 tracks. Derive tracks from the argument + existing gaps.
Apply
protocols/decision.mdDISAMBIGUATION if topic is ambiguous. - Assign output shape per track (API contracts, pattern comparison, claim verification, domain survey). If no topic provided, infer from recent conversation context.
Research (HIGH freedom):
Delegate to specwright-researcher per protocols/delegation.md. One call per track.
Parallel if Agent Teams available. Cite all sources with URLs. UNFETCHED → noted, not fabricated.
Synthesis (MEDIUM freedom): Merge findings. Score confidence: HIGH (official docs, multiple sources), MEDIUM (reputable secondary), LOW (single, unverified). Tag LOW/MEDIUM as potential assumptions. Flag open questions honestly.
Auto-approval (MEDIUM freedom): If all tracks have ≥MEDIUM confidence: auto-approve and persist the brief. Low-confidence tracks noted in the brief for consumer awareness. Briefs are consumed by sw-design which validates findings at its own gate.
Persistence (LOW freedom):
Write to {projectArtifactsRoot}/research/{topic-id}-{YYYYMMDD}.md. Overwrite same topic+date.
Max 10 briefs. If at cap: log warning, list by date, suggest cleanup.
Briefs older than 90 days are STALE — warn when loading.
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.
- 6d ago First seen · 82 lines · 29 tokens per session scan A 2f67b30e7d0a
sw-research is a skill published in the GitHub repository Obsidian-Owl/specwright (9 stars, last pushed 4mo ago), licensed MIT. It adds 29 tokens to every session and 742 once invoked, about $0.0001 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 skills, from other repositories
rust-skills
Rust best practices — 179 rules across 14 categories for idiomatic, optimized Rust code.
new-skill
Scaffold a new brooks-lint analysis skill so it passes npm run validate and npm run evals on the first try — generates skills/{name}/SKILL.md (with the mandatory "Do NOT trigger for:" clause and a Process section citing guide step ranges) plus skills/{name}/{name}-guide.md (sequentially numbered steps), then appends…
llm-friendly-context
Clarifies inputs, outputs, success criteria, decisions, and unresolved conditions so downstream consumers can execute without guessing. Use when writing or revising LLM-facing prompts, handoffs, planning artifacts, reviews, reports, or generated instructions.
import-spec
Import a spec-kit feature spec into Shipyard, replacing brainstorming. Use when a spec-kit feature directory exists with spec.md.
worktree
Isolate work in a git worktree. Activates when the user needs branch isolation before executing a plan, wants a clean environment for a feature, is ready to finish work in an existing worktree, or wants to reclaim worktrees that accumulated and are now polluting filesystem-level measurement.
write-article
Write high-quality articles from scratch using a 6-phase pipeline: deep STORM-inspired research with parallel agents, multi-perspective outline generation, section-by-section drafting with research grounding, adaptive anti-slop enforcement (hard/soft banned vocabulary per tone), cross-model adversarial review, and SEO…