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 agentmods add commands/me2resh/agent-decision-record/decidegit clone --depth 1 https://github.com/me2resh/agent-decision-recordWhat 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 | $0.00000 | $0.00826 |
| Opus 5 | $0.00000 | $0.00413 |
| Sonnet 5 | $0.00000 | $0.00165 |
| Haiku 4.5 | $0.00000 | $0.00083 |
Grade C, and why
decide scanned grade C 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.
Hidden instructionshighPrompt injection
Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.
<!-- Canonical source for the /decide process. skills/decide/SKILL.md mirrors the Trigger/Process/Rules sections below (Claude Code skills need self-contained instructions and can't just point at another file) — keep the The source is not reproduced here
No licence file
A repository with no LICENSE is all rights reserved by default, so the body is not copied here. The metadata, the measurements and the link are.
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.
- 2d ago First seen · 134 lines · 0 tokens per session scan C 60cfcc1d43a8
decide is a command published in the GitHub repository me2resh/agent-decision-record (44 stars, last pushed 25d ago), with no licence file. It costs nothing until one of its globs matches a file; then it loads 826 tokens. A static security scan graded it C with 1 finding (hidden instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
adr-backfill
Audit code, documentation, and history for potential architecture decisions that were never recorded, then return a deduplicated evidence report. Read-only.
adr-draft
Draft one new ADR from a current decision or a selected backfill handoff. Writes a single new record, as proposed.
adr-check
Check the current change, or a plan, against the decisions that govern it and report a verdict per decision. Read-only.
adr-queue
Show the review queue — every decision still proposed, with its SLA state — so open questions are visible before they are planned around. Read-only.
adr-context
Load the architecture decisions that govern the paths you are about to change, before planning against them. Read-only.
check
Check a produced plan against the decisions that govern it. Read-only.