Borrowing it
Nothing to install: this file belongs to JJHbrams/Project-AMBER. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/JJHbrams/Project-AMBER/master/.github/skills/engram-hook-trust/SKILL.mdgit clone --depth 1 https://github.com/JJHbrams/Project-AMBERWrote 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/skills/jjhbrams/project-amber/engram-hook-trust)<a href="https://agentmods.dev/skills/jjhbrams/project-amber/engram-hook-trust"><img src="https://agentmods.dev/badge/skills/jjhbrams/project-amber/engram-hook-trust/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/skills/jjhbrams/project-amber/engram-hook-trust"><img src="https://agentmods.dev/badge/skills/jjhbrams/project-amber/engram-hook-trust.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 | $0.00741 |
| Opus 5.5 | $0.00013 | $0.00296 |
| Sonnet 5.5 | $0.00007 | $0.00148 |
| Haiku 4.5 | $0.00003 | $0.00074 |
Grade A, and why
engram-hook-trust 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 1mo 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.
What it actually says
Engram Hook Trust
Use this skill only for the native Codex engram_report_codex_event MCP hooks.
It covers exactly: UserPromptSubmit, PreToolUse, PostToolUse,
PermissionRequest, Stop, Interrupt, SubagentStart, and SubagentStop.
It never manages SessionStart, titles, command hooks, arbitrary servers, or
other approval state.
Run scripts/engram-hook-trust.mjs --help first. The script uses Codex
app-server APIs and does not parse or rewrite TOML.
checkis read-only and reports finite status/reason codes. With--backup, it also reports how many snapshot entries are currently recoverable.backupis read-only against Codex. It creates a new immutable backup of the trusted-and-enabled subset only; it never adds an untrusted event.restorerequires--confirm --backup <file>. It restores only a missing state record whose live canonical key and generated native definition hash exactly match the backup. It skips already trusted/enabled state and rejects existing untrusted/disabled state, changed definitions, unknown status, or root-mismatched state. An absent record reported asuntrustedis eligible only when its unchanged approval was captured in this root's backup.
Never use a backup to manufacture approval for a changed definition, another
Codex root, or any hook outside this eight-event allowlist. Do not run restore
unless the user explicitly requests that specific restore operation.
Show the check result first: deletion alone cannot distinguish Orca's cleanup
from intentional user revocation. Never watch/restore automatically or import
approvals from config.toml.bak. If an approval disappeared before a valid
backup existed, ask the user to review it in Codex /hooks, then make a new
backup. Do not run a provider conversation or event hook to perform these tasks.
Typical invocation requires the native Codex executable, not the codex.cmd
npm shim: node scripts/engram-hook-trust.mjs check --root "$env:CODEX_HOME" --codex-executable <path-to-codex.exe>. Backups must live under
$env:CODEX_HOME/.engram-hook-trust/.
Resolve the installed native executable and the target root explicitly; do not
mix the ordinary Codex home and Orca home. Run from the skill directory, or use
the absolute script path. Use a new filename for every backup. For example,
after setting task-local $trustScript, $trustRoot, $trustExe, and
$trustBackup to those verified paths:
node $trustScript backup --root $trustRoot --backup $trustBackup --codex-executable $trustExe
node $trustScript check --root $trustRoot --backup $trustBackup --codex-executable $trustExe
# Only after the user requests this specific recovery:
node $trustScript restore --root $trustRoot --backup $trustBackup --codex-executable $trustExe --confirm
Report partial backup counts honestly. A backup with two records cannot restore
the other six hooks. A failed post-write verification may mean a write occurred;
inspect with check and stop instead of retrying or rolling back approvals.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago First seen · 58 lines · 33 tokens per session scan A abceafc4eea9
engram-hook-trust is a skill published in the GitHub repository JJHbrams/Project-AMBER (2 stars, last pushed 14d ago), licensed MIT. It adds 33 tokens to every session and 741 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-09-13.
Other skills, from other repositories
penfield
Persistent memory for AI agents. Store decisions, preferences, and context that survive across sessions. Build knowledge graphs that compound over time. Hybrid search (BM25 + vector + graph) recalls what matters when you need it.
memory-audit-discoverability
A review guide for checking whether stored memories can be found at the right time. It focuses on where memories are attached, when they are triggered, whether aliases are missing, and whether a parent has too many children.
memory-audit-pattern-extraction
A method for investigating repeated mistakes by comparing related memories and checking whether an earlier reminder failed. It looks at where the reminder was stored, when it was created, and whether it was strong enough to prevent the mistake.
memory-audit
An entry point for reviewing and maintaining an AI agent's stored memories. It describes how to remove repetition, preserve useful reasoning, and update memories when old conclusions no longer fit.
add-connector
Onboard a deus connect connector — a way to route Claude Code sessions to non-Claude models (e.g. GPT via CLIProxyAPI) alongside normal Claude access. Generic orchestrator over whichever connector is selected.
mengram-memory
Long-term memory with 3 types (facts, events, workflows). Remember user preferences, past conversations, and learned procedures across sessions. Use when recalling what the user said before, saving important info, getting user context, or tracking completed workflows.