Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/roomi-fields/paper-trailnpx agentmods add commands/roomi-fields/paper-trail/paper-trail-reviewWrote 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/roomi-fields/paper-trail/paper-trail-review)<a href="https://agentmods.dev/commands/roomi-fields/paper-trail/paper-trail-review"><img src="https://agentmods.dev/badge/commands/roomi-fields/paper-trail/paper-trail-review/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/roomi-fields/paper-trail/paper-trail-review"><img src="https://agentmods.dev/badge/commands/roomi-fields/paper-trail/paper-trail-review.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 | $0.00039 | $0.01039 |
| Opus 5 | $0.00019 | $0.00519 |
| Sonnet 5 | $0.00008 | $0.00208 |
| Haiku 4.5 | $0.00004 | $0.00104 |
Grade A, and why
paper-trail-review 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 — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/paper-trail:review — Arbitrate problematic refs
Pour traiter en une session toutes les refs qui ne peuvent pas progresser automatiquement et demandent une décision humaine.
Usage
/paper-trail:review # tous les cas
/paper-trail:review --auto-retract-uncited # raccourci : retract en lot les non-citées
/paper-trail:review --only retract # seulement les artefacts évidents
/paper-trail:review --only blocked # seulement les paywalls
/paper-trail:review --only investigate # seulement les homonymies
/paper-trail:review --batch # propose une décision par ref, l'utilisateur valide en lot
Raccourci --auto-retract-uncited
Empiriquement, 90 %+ des refs problématiques sont des refs non citées en dehors du registre. Pour les traiter en une commande :
python3 -m pipeline retract-uncited # dry-run (montre la liste)
python3 -m pipeline retract-uncited --apply # exécute les retract
Le pipeline scanne le vault, identifie les refs actives non citées hors
_registry/INDEX.md, et propose de toutes les retract en bloc.
Ce que fait Claude pendant cette commande
-
Régénère le rapport à jour :
python3 tools/review_problems.py \ --states candidate,uid_resolved,awaiting_rtfm_ocr \ --output plans/review_<YYYY-MM-DD>.md -
Lit le rapport et compte par catégorie (RETRACT / BLOCKED_HUMAN / INVESTIGATE / WAIT / REVIEW). Présente le total à l'utilisateur.
-
Traite par batch, dans cet ordre :
- D'abord les RETRACT évidents (title vide ou "untitled", uid absent,
citée nulle part ou seulement dans
INDEX.md) → propose une liste en bloc, demande validation globale. - Ensuite les BLOCKED_HUMAN (auteur+titre OK, ≥8 sources tentées, toutes en échec) → propose une liste en bloc.
- Enfin les INVESTIGATE : un par un, car nécessite une décision d'humain (homonymie ? typo dans frontmatter ?).
- D'abord les RETRACT évidents (title vide ou "untitled", uid absent,
citée nulle part ou seulement dans
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 · 97 lines · 39 tokens per session scan A 84a056fe3b02
paper-trail-review is a command published in the GitHub repository roomi-fields/paper-trail (5 stars, last pushed 7d ago), licensed MIT. It adds 39 tokens to every session and 1,039 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
gh-security-review
Use to perform a dedicated security review of current branch changes before pushing.
gh-release
Create a new release for the marketplace or a specific plugin.
ledger-research
Launch parallel evidence collection across all active research pillars.
ledger-spec
Generate constrained PRD and architecture documents where every section cites decisions.
quick-research
Quick research with source verification (lighter than /deep-research).
gh-status
Quick overview of current GitHub workflow state.