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.
git clone --depth 1 https://github.com/ethanaubuchon/dossier-tradecraftWrote 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/commands/ethanaubuchon/dossier-tradecraft/decompose)<a href="https://agentmods.dev/commands/ethanaubuchon/dossier-tradecraft/decompose"><img src="https://agentmods.dev/badge/commands/ethanaubuchon/dossier-tradecraft/decompose/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/commands/ethanaubuchon/dossier-tradecraft/decompose"><img src="https://agentmods.dev/badge/commands/ethanaubuchon/dossier-tradecraft/decompose.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 | $0.00033 | $0.02927 |
| Opus 5 | $0.00016 | $0.01463 |
| Sonnet 5 | $0.00007 | $0.00585 |
| Haiku 4.5 | $0.00003 | $0.00293 |
Grade A, and why
decompose 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 10d 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 — 138 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are starting a /decompose session. The topic (if provided): $ARGUMENTS
This is the /decompose recipe of the dossier-tradecraft framework — Phase 3, feature/story breakdown. Consumes scope docs at status: ready-for-decompose and produces a task graph on the project's issue tracker.
Principles
- Tracker is canonical; vault is breadcrumb. Tasks live in the issue tracker. The scope doc gets a
## Tracked atappend pointing at the issues; no new vault artifact is created. Tracker-as-canonical eliminates the drift between PR auto-close and vault checkboxes. - Two-track behavior. Fast-track for 1–2 issues (small bug fix, single feature). Standard-track for 3+ (sizing + dependency graph; layered fallback for larger breakdowns). Same recipe, different ceremony.
- The graph is the value. Sizing and dependencies are coupled — slicing tasks naively loses the parallelism reasoning an LLM is well-suited to do. Don't enumerate; graph.
- Project-adaptive shape. Tracker preferences are read from the project's agent-instruction file (CLAUDE.md, AGENTS.md, etc.) (or other project context) as prose — no schema. Vault fallback when no tracker is configured.
- Cross-team artifact self-containment. Issue bodies do not reference the vault — issue trackers may be accessed by collaborators without vault access, and vault references are dead-end for them. Vault → tracker links exist one-way only, on the vault side.
- Agent-judged writes. Decide sizing call, structure, and prose directly; surface what was done. The only explicit confirmation gate is the fast-track sizing check.
- Loop exits on natural signal. No 3-option decision matrix at every cycle.
Workflow
- Load context. Invoke the
load-contextskill (vault grounding: profile, inbox check, project area pre-warm). If the working directory is inside a git repo, also note the cwd, current branch, and whether a project agent-instruction file (CLAUDE.md, AGENTS.md, etc.) is present, as additional context for the session.
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.
- 10d ago First seen · 138 lines · 33 tokens per session scan A 64cc1fd58dfb
decompose is a command published in the GitHub repository ethanaubuchon/dossier-tradecraft (3 stars, last pushed 2mo ago), licensed MIT. It adds 33 tokens to every session and 2,927 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.
Other commands, from other repositories
session-end
Thin wrapper (GETSITREP-10) over getsitrep session-end — the CLI owns the Session Log entry, header updates, .sitrep-data.json, the history record, and the commit. This file's only remaining job is the one step no CLI can do on its own: building the session-summary JSON from this conversation, since a CLI has no way…
capture
Capture a new task into the project plan and status report in one step. Never lose an idea mid-session.
sitrep
Thin wrapper (GETSITREP-10) — reading STATUSREPORT.md, the session-command tracker (absorbed from the old /pulse, see docs/specs/command-canon.md), and the next-command suggestion logic all live in getsitrep sitrep.
selfaware-task-update
Create or update TASKSTATE.md in the project root with the current session's progress (manual checkpoint).
flow-status
Show the current cairnkit run — stage, history, pending approvals, artifacts.
para-status
Show an overview of your PARA system — item counts, statuses, and attention-needed flags.