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 fledgeling-co/fledgeling-plugins --skill trawlgit clone --depth 1 https://github.com/fledgeling-co/fledgeling-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/skills/fledgeling-co/fledgeling-plugins/trawl)<a href="https://agentmods.dev/skills/fledgeling-co/fledgeling-plugins/trawl"><img src="https://agentmods.dev/badge/skills/fledgeling-co/fledgeling-plugins/trawl/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/fledgeling-co/fledgeling-plugins/trawl"><img src="https://agentmods.dev/badge/skills/fledgeling-co/fledgeling-plugins/trawl.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 | $0.00218 | $0.04100 |
| Opus 5.5 | $0.00087 | $0.01640 |
| Sonnet 5.5 | $0.00044 | $0.00820 |
| Haiku 4.5 | $0.00022 | $0.00410 |
Grade A, and why
trawl 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 — 308 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Trawl — cast wide, sift hard, ship the catch
The first three answers the model gives are the answers a senior engineer gives in thirty seconds. Correct. Forgettable. Worse: they are the same answers every time — when researchers over-sampled 4,000 ideas from one aligned model, ~95% were semantic duplicates of each other. Diversity does not come from asking again, from raising temperature, or from telling the model to "be creative". It comes from architecture: isolated parallel frames, explicit differentiation, and convergence rules that protect distinct mechanisms instead of rewarding the highest average score.
This skill is that architecture. It also fixes the two known failure modes of naive divergent ideation: recommending a dazzling idea that doesn't solve the stated problem (defended by the baseline boss-gate), and wasting branches on frames that don't fit the problem (defended by the frame portfolio and apoptosis).
references/evidence.md holds the research grounding with citations —
read it when you need to justify or tune a mechanism, not on every run.
Running as a Gemini model? Read gemini.md in this directory first, then follow this file with the overrides it names. It writes the frozen baseline, the frame ledger and each branch's ideas to files that later phases read back, computes every receipt number from those files, and reads each stated bound off the produced set. Other models skip it.
Pre-flight router (run before anything else)
This skill is expensive: roughly 8–14 Agent calls at the standard tier, 5–10x a single answer. The router decides the tier; it never silently skips the run when the user opted in.
Step 1 — explicit invocation. If the user typed /trawl:trawl (or legacy /adhd / "ADHD mode"), run it. --any, --standard, --100 set the tier
directly; with no flag, default to standard. Do not second-guess an
explicit opt-in.
Step 2 — self-judge (only when Step 1 didn't match). Three questions:
- Open-ended? Would a senior engineer give multiple viable answers, or is there one canonical answer? Canonical → answer directly.
- Stakes? High (architecture, public API, naming a real product, fuzzy bug with no known root cause, schema design) → standard tier. Medium (real decision, low blast radius) → any% tier. Trivial → answer directly.
- Open phrasing? "Quick", "standard", "canonical", "textbook",
"just", "one-line" mean the user wants the direct answer. Answer
directly, optionally appending: "For a wider exploration under
parallel frames with trap detection, run
/trawl:trawl <problem>."
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 Changed · +3 tokens per session 042aef50177d
- 1mo ago First seen · 308 lines · 215 tokens per session scan A 3aff773df1e4
trawl is a skill published in the GitHub repository fledgeling-co/fledgeling-plugins (2 stars, last pushed 4d ago), licensed MIT. It adds 218 tokens to every session and 4,100 once invoked, about $0.0009 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 2026-09-04.
Other skills, from other repositories
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
chat-pet-sprite-creation
Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…