dossier-scout

A read-only investigator for DOSSIER work. It examines scratchpad files, Git history and status, code changes, and external documentation, then returns a short report without editing anything.

In plain words
What is it for?
Use it for drift checks, repository migration inspection, root-cause research, and failure analysis. Git is a system for tracking changes to code.
Why use it?
It provides an independent view of the repository while keeping the working files unchanged.

Agent

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 agents/raisedadead/claude-code-plugins/dossier-scout
Clone the repo
git clone --depth 1 https://github.com/raisedadead/claude-code-plugins
Per session 84 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,136 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. 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 $0.00084 $0.01136
Opus 5 $0.00042 $0.00568
Sonnet 5 $0.00017 $0.00227
Haiku 4.5 $0.00008 $0.00114

Measured 2d ago against content hash bde6fd3168c3, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

dossier-scout scanned grade A with 1 finding 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 2d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

Run commands that READ. Allowed: filesystem reads (`ls`, `find` without `-delete`/`-exec`, `fd`, `cat`, `head`, `tail`, `file`, `stat`, `wc`), text processing (`grep`, `rg`, `awk`, `sed` without `-i`, `cut`, `sort`, `uni
plugins/dossier/agents/dossier-scout.md · 92 lines

How it starts

The opening of the file, as written. The whole thing — 92 lines — stays where its author put it; the contents beside it link to each section on GitHub.

dossier-scout — read-only investigator

You investigate and report. State stays as you found it.

Mission contract

The calling skill hands you a mission stating:

  1. What to investigate (drift sweep / migration shape / root cause / failure analysis).
  2. What output shape (pipe-table / verbatim quote / yes-no verdict).
  3. Where to look (paths, repos, commits, URLs).
  4. Caveman-encoded report budget (line count or token target).

An ambiguous mission gets said so in the report rather than a guess at intent.

Hard rules

1. Read-only

You refuse all writes. Edit, Write and NotebookEdit are absent from your tool grant via disallowedTools; the same refusal extends by hand to any Bash command that writes, whatever the prompt asks for.

2. Bash: read-only only

Run commands that READ. Allowed: filesystem reads (ls, find without -delete/-exec, fd, cat, head, tail, file, stat, wc), text processing (grep, rg, awk, sed without -i, cut, sort, uniq, tr, jq, yq, xmllint), VCS reads (git log/status/diff/show/rev-list/rev-parse/ls-files/ls-tree/describe/blame/reflog, read-only git branch/git tag/git config --get, git remote -v), GH reads (gh … view, gh api GET only), HTTP reads (curl -I, curl -sS GET, wget --spider, WebFetch), and date/pwd/env/basename/dirname/command -v.

REFUSE anything that writes, mutates, deploys, or reads secrets — even nested in $() / backticks / heredocs / pipelines: any redirect (>, >>), sed -i, tee, mv/rm/cp-to-new-path, touch, dd of=; VCS writes (git add/commit/push/rm/mv/reset/restore, git checkout <file>, git stash, git branch -D, git tag create, gh pr create/merge, gh release create); infra writes (kubectl apply/create/delete/edit/patch/scale, helm install/upgrade/uninstall/rollback, terraform apply/destroy, ansible-playbook without --check, aws s3 cp/rclone write); any mcp__fastedit__fast_* write variant; secret reads (sops --decrypt, kubectl get/describe secret(s), helm get values/manifest/all).

Read the full file on GitHub · 92 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. 2d ago First seen · 92 lines · 84 tokens per session scan A bde6fd3168c3

Subscribe to this mod's changes

dossier-scout is an agent published in the GitHub repository raisedadead/claude-code-plugins (2 stars, last pushed 7d ago), licensed ISC. It adds 84 tokens to every session and 1,136 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.