shadow-reference

shadow-reference is a skill for Claude Code, Codex from michal-bajer1/shadow. It costs 48 tokens per session (2,154 once invoked), scanned A, original, MIT.

A reference guide for shadow-cli, a command-line tool for recording implementation decisions and identifying drift.

In plain words
What is it for?
Use it when working with shadow-cli, interpreting its .shadow files, resolving drift, or creating decision records.
Why use it?
It gives agents the exact commands and rules needed to create, read, and update decision records consistently.

Skill for Claude CodeCodex

Part of the shadow plugin — 4 skills, 3 hooks shipped together

Install

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.

agentmods
npx agentmods add skills/michal-bajer1/shadow/shadow-reference
Any agent
npx skills add michal-bajer1/shadow --skill shadow-reference
Clone the repo
git clone --depth 1 https://github.com/michal-bajer1/shadow

Made for: Claude Code, Codex.

Or install shadow, the plugin that ships this one along with the rest of its 4 skills, 3 hooks.

Wrote 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.

agentmods badge for shadow-reference

README.md
[![agentmods](https://agentmods.dev/badge/skills/michal-bajer1/shadow/shadow-reference.svg)](https://agentmods.dev/skills/michal-bajer1/shadow/shadow-reference)
Your own site
<a href="https://agentmods.dev/skills/michal-bajer1/shadow/shadow-reference"><img src="https://agentmods.dev/badge/skills/michal-bajer1/shadow/shadow-reference.svg" alt="Measured on agentmods" height="20"></a>
Per session 48 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,154 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00048 $0.02154
Opus 5 $0.00024 $0.01077
Sonnet 5 $0.00010 $0.00431
Haiku 4.5 $0.00005 $0.00215

Measured 3d ago against content hash eb4b3ecb53fc, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

shadow-reference 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 3d 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.

plugins/shadow/skills/shadow-reference/SKILL.md · 224 lines

How it starts

The opening of the file, as written. The whole thing — 224 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Commands

record

shadow-cli record \
  --paths <file> [<file> ...]   files where the decision lives \
  --dirs  <dir>  [<dir> ...]    directories when the decision spans a module \
  --task        "..."           the initiative or mandate — what work this belongs to \
  --constraints "..." "..."     rules or facts that shaped your choices \
  --decision    "..."           the specific implementation choice and why — must not restate the task \
  [--rejected   "..." "..."]    alternatives considered, each with why discarded \
  [--confidence high|medium|low] \
  [--forward    "..."]          unresolved thread a future agent will need to complete \
  [--supersedes <uuid>]         replace a prior record

Use --paths for the files that express the decision (core logic, resource declaration, config). Exclude files that only reference or test it.

--task and --decision answer different questions. --task is the retrieval anchor — "which initiative did this change belong to?" --decision is the reasoning — "within that initiative, what specific choice was made and why?" If --decision could serve as a one-line summary of --task, it is too vague: it is restating the goal rather than explaining the implementation choice. Use --dirs instead of --paths when the decision is about the directory as a whole (a module, a service). Use both when the decision is file-specific but you also want the directory tracked for drift.

query

shadow-cli query <path>                decisions for the nearest .shadow/ ancestor
shadow-cli query --all <path>          walk up from path to repo root, return all
shadow-cli query --exact <path>        only records whose blob SHA still matches (no drift)
shadow-cli query --module <dir>        only records whose tree SHA still matches
shadow-cli query --unconfirmed <path>  only records that have not yet been confirmed

Use --all before making changes — it returns decisions from the file's directory AND all parent directories up to repo root. Use --exact when you need to confirm a record is still current. Use --module to find decisions that still apply to the whole directory tree. Use --unconfirmed to find records that still need review.

Read the full file on GitHub · 224 lines

Changes

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.

  1. 3d ago First seen · 224 lines · 48 tokens per session scan A eb4b3ecb53fc

Subscribe to this mod's changes

shadow-reference is a skill published in the GitHub repository michal-bajer1/shadow (2 stars, last pushed 3mo ago), licensed MIT. It adds 48 tokens to every session and 2,154 once invoked, about $0.0002 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.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.

obra/superpowers · 37 tokens

auto-perf-optimize

Run agent-driven VS Code performance or memory investigations. Use when asked to launch Code OSS, automate a VS Code scenario, run the Chat memory smoke runner, capture renderer heap snapshots, take workflow screenshots, compare run summaries, or drive a repeatable scenario before heap-snapshot analysis.

microsoft/vscode · 62 tokens

chat-perf

Run chat perf benchmarks and memory leak checks against the local dev build or any published VS Code version. Use when investigating chat rendering regressions, validating perf-sensitive changes to chat UI, or checking for memory leaks in the chat response pipeline.

microsoft/vscode · 51 tokens

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.

microsoft/vscode · 53 tokens

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…

microsoft/vscode · 71 tokens