ShadowFrog is a suite of coding-agent skills that maintains a file-backed knowledge base of tacit information about a codebase, such as fragile dependencies, untested invariants, and hard-to-notice edge cases. It helps coding agents preserve lessons from code reading, experiments, and conversations across future work. The catalogue contains skills that create, update, explore, organize, and view this shadow knowledge base.
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 microsoft/ShadowFrog --skill shadow-frog-napgit clone --depth 1 https://github.com/microsoft/ShadowFrogWrote 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/microsoft/shadowfrog/shadow-frog-nap)<a href="https://agentmods.dev/skills/microsoft/shadowfrog/shadow-frog-nap"><img src="https://agentmods.dev/badge/skills/microsoft/shadowfrog/shadow-frog-nap/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/microsoft/shadowfrog/shadow-frog-nap"><img src="https://agentmods.dev/badge/skills/microsoft/shadowfrog/shadow-frog-nap.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 | $0.00106 | $0.04961 |
| Opus 5.5 | $0.00042 | $0.01984 |
| Sonnet 5 | $0.00021 | $0.00992 |
| Haiku 4.5 | $0.00011 | $0.00496 |
Grade A, and why
shadow-frog-nap 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 — 472 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ShadowFrog Nap
Nap is implementation-free feature-task ideation, not feature development or verified dreaming. The host agent generates proposals and delegates an independent judgment. The Python helper manages the persistent tree, binds judgments to exact proposal inputs, and exports reviewed contracts. It never calls a model, executes probes, implements features, creates worktrees/branches, pushes, or writes shadow discoveries.
Inputs and Limits
Accept a user goal or area, optional prior dream/nap evidence, a mode, and a
work budget. A Git repository with a commit is required; a remote and an
initialized .shadow/ are not required.
| Setting | Default | Meaning |
|---|---|---|
mode |
broad |
Explore distinct opportunities; coherent regularizes parent-child connections |
max_nodes |
7 | Total recorded nodes, including imported seeds and rejected attempts |
max_depth |
unset (null) |
Optional user-requested maximum parent edges; roots have depth 0 |
max_probes |
2 | Total executed probes recorded across all nodes, including failures |
max_tasks |
2 | Maximum selected ready task briefs |
max_reviews |
2 | Recorded judge batches: one shortlist review plus a bounded re-review |
These are ceilings, not quotas. Zero selected tasks is a valid result.
Users can request larger budgets, including ten diverse children of one
parent. Agree on limits before starting and persist them in the record;
never erase rejected work or enlarge limits to disguise an overrun.
Depth is not a default stopping condition. Omit max_depth or set it to
null to leave it uncapped; an explicit nonnegative integer still limits it.
The total node budget bounds every run, and parent/cycle checks always apply.
The helper enforces recorded limits, not actual API spending or unrecorded commands/model calls. Respect the host's token/cost/time limits separately, including invalid or failed judge responses. Do not install dependencies, debug infrastructure, spawn per-candidate implementation workers, or run full suites automatically. One independent shortlist judge and a bounded re-review are allowed within the review budget; do not start a model panel for every node. Escalation to a full dream is a separate decision.
What ships with it
1 file 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.
- 6d ago First seen · 472 lines · 106 tokens per session scan A de913cd202cb
shadow-frog-nap is a skill published in the GitHub repository microsoft/ShadowFrog (28 stars, last pushed 4d ago), licensed MIT. It adds 106 tokens to every session and 4,961 once invoked, about $0.0004 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 2026-09-23.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
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…
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…